Photocatalytic Performances and Antifouling Efficacies of Alternative Marine Coatings Derived from Polymer/Metal Oxides (WO₃@TiO₂)-Based Composites

Abstract

This work concerns development of alternative antifouling paints for marine applications using composite metal oxides derived from TiO₂ and WO₃. Composite metal oxides with a variety of tungsten content were prepared via a sol–gel process using titanium isopropoxide and sodium tungstate dihydrate as the precursors. The crystalline phase, bandgap energy, morphology, surface structure, and electronic states of the synthesized products were then characterized and confirmed by XPS, XRD, UV/Vis spectroscopy, SEM-EDX, and TEM techniques. Photocatalytic performance polymer film loaded with composite metal oxides containing 10% by mole of WO₃ (10%WO₃@TiO₂) was confirmed both under UV irradiation and in the dark. The results are discussed in light of oxygen vacancies and the presence of heterojunctions between the TiO₂ and WO₃ domains in the composites, which eventually lead to suppression of charges recombination. Finally, antifouling and the antimicrobial efficacy of the polymer film loaded with composite metal oxide particles (10%WO₃@TiO₂) were evaluated under static marine immersion conditions using Zobell Marine agar. After 30 days, the percentage fouling coverage (16.35%), colonies number (CFU value 12 × 10³), and percentage reduction of colonies (92.94%), were obtained, which significantly outperformed those of the control (the bare substrate).

1. Introduction

Marine biofouling has long been a serious problem in many marine applications, including ship hulls and other submerged structures, such as pipes, gates, and stationary structures. The problem is also associated with both environmental and economic viewpoints. For instance, the presence of barnacles on a ship’s hull could lead to higher frictional resistance and greater fuel consumption for the ship. Fuel consumption by waterborne transportation has been estimated at approximately 300 million tons per year and greater consumption is forecast for the future. The increase in fuel consumption also contributes to greenhouse gas emissions.

Biofouling on a ship can also lead to introduction of invasive species. Molnar et al. [1] found that 39% of the 329 invasive species were introduced through shipping via hull fouling. Specifically, in Port Philip Bay, Australia, biofouling was responsible for 78% of marine species casually introduced into the region [2]. Invasive species can cause ecological, environmental, and economic damage by, for example, altering ecosystems and trophic chains through displacement of endemic species. Invasive species can also cause pathogenic effects, leading to decreased populations of microorganisms in the region. Furthermore, invasive species can also compete with fishing species, reducing fishing production rates. Moreover, hard-shelled barnacles can degrade paint, leading to other problems, such as metal corrosion.

To cope with the above problems, one possible strategy includes use of antifouling paints. In general, antifouling paints can be classified into two main classes: non-toxic coatings and biocidal coatings [3]. The former system relies on a fouling release mechanism. Examples of these coatings include polytetrafluoroethylene (Teflon)- and silicone-based coating systems. By contrast, the latter coating system relies on use of active ingredients, such as antifouling biocides and metallic-based materials. In the past, biocidal coatings derived from heavy metal compounds, such as mercury and arsenic, were used as active antifouling agents. However, these chemicals are toxic and have become obsolete. Likewise, use of antifouling paints containing tributyltin (TBT) has also been restricted by the Environmental Protection Agency (EPA), USA [4]. In Europe, use of underwater antifouling paints containing TBT on ships of 400 GT and above has been prohibited by European law (EC Regulation No. 782/2003). In Australia, metallic copper, cuprous oxide, thiuram, and zinc oxide have been included in the list of approved antifouling biocides, whereas other chemicals, such as triazine derivative and benzothiazole derivative (TCMTB), have yet to be approved [5].

Additionally, the bactericidal efficacy of TiO₂-based photocatalysts has been reported [6]. This is related to formation of active oxygenic species, such as hydrogen peroxides and hydroxy radicals, which can damage cell membranes of bacteria. Eventually, the above effect leads to loss of function, an increased rate of mutagenesis, and ultimately cell death [7]. However, the photocatalytic activity of TiO₂ must be activated by UV irradiation. This limits use of the TiO₂ photocatalyst under more versatile light illumination conditions, such as under visible light (for indoor air quality control) and/or in the dark. To enhance its performance, the TiO₂ photocatalyst has been coupled with different types of metal oxides [8]. These include sol–gel-synthesized Fe₂O₃/TiO₂ and Y₂O₃/Fe₂O₃/TiO₂ [9], ball milling and sol–gel-synthesized ZnO/TiO₂ [10,11], in situ polymerized polypyrrole/TiO₂/(TiO₂–V₂O₅) composite particles [12], and TiO₂/WO₃ composite metal oxides [13,14]. Due to the unique properties of WO₃, including its wide bandgap, good electrochemical performance, good redox properties, and versatility of the preparation methods [15], in this study, use of WO₃ for coupling with TiO₂ was of interest. Ismail et al. [16] prepared mesoporous WO₃–TiO₂ composites containing different concentrations of WO₃ (0–5 wt%) via the sol–gel process. From a photocatalytic degradation test against imazapyr, it was found that the efficiency of the composite containing 3 wt% of WO₃ was 3.5 times greater than that of the normal mesoporous TiO₂ (without coupling with WO₃). Moreover, when the test was carried out under visible light, the composite metal oxide containing 0.5 wt%WO₃ was the best formulation, providing 46% of the photocatalytic efficiency. The result relates to the decrease in the bandgap energy value of TiO₂ after coupling with WO₃. Likewise, Riboni et al. [17] tried to suppress recombination of electron–hole generated from photocatalytic reaction of TiO₂ by coupling with WO₃ using two types of precursors—sodium tungstate (Na₂WO₄) and tungsten methoxide (W(C₂H₅O)₆. Greater photocatalytic activity of the composite metal oxides (tested against formic acid and acetaldehyde under UV irradiation) was achieved when the percentage loading of the tungsten precursor was 3% by mole. It was also found that the performance of the composite metal oxides prepared using the organic precursor (tungsten alkoxide) was greater than that prepared using sodium tungstate. The discrepancy was ascribed to the different morphology and crystalline structure of the products.

Tatsuma et al. [18] studied the corrosion resistance of 304-type stainless steel substrates coated with different types of TiO₂-based films. Interestingly, it was demonstrated that the anticorrosion efficacy of the films derived from TiO₂ coupled with WO₃ could be maintained even after the UV light was turned off. The above effect was ascribed to the role of the WO₃ phase, which acted as an electron pool or an energy storage substance, and could be discharged in the dark. Subsequently, Tatsuma et al. [19] showed that ITO electrodes coated with TiO₂–WO₃ films exhibited a photochromism effect, both in water and in the air, provided that the relative humidity was sufficiently high (above 25%). This was not the case for the ITO coated with neat WO₃ film. The concept of an energy storage substance in the system was also demonstrated and discussed in light of an intercalation of the WO₃ phase by the protons (H⁺) generated from the photocatalytic reactions. It was also recommended that interfacial contact between the TiO₂ and WO₃ domains must be sufficiently strong to facilitate charge transfer between the TiO₂ and WO₃ in the system. Jaritkaun et al. [20] prepared mixed metal oxides via simple mixing of TiO₂ particles with WO₃ particles. To enhance the interfacial adhesion between the two metal oxide domains, an in-situ-polymerized thiophene was integrated into the system. It was found that mixed metal oxides (TiO₂/WO₃/PTh) with a weight ratio of 1/2/0.4 provided the best composition, capable of maintaining degradation of methylene blue in the dark. Additionally, the concept of introducing an energy storage substance into metal oxide systems was also applied to enhance the bactericidal performance of the TiO₂. US patent 7989520B2 [21] disclosed an invention related to antifouling compositions free from organotin and cuprous oxide compounds. The main composition of the TBT free paint includes self-polishing zinc acrylate polymer mixed with alkaline metals, alkaline earth metals, metal oxides (MgO, TiO₂, ZnO), and silicate mineral (tourmaline). Tatsuma et al. [22] investigated the bactericidal effects on E.Coli using TiO₂–WO₃ composite films prepared by using crystalline WO₃ nanoparticles and bis(2,4-pentanedionato) titanium (IV) oxide as the raw materials. After charging under UV radiation, the antibacterial performance of the composite metal oxides films was tested in the dark and the efficacy could be maintained for 6 h before the surviving bacteria were killed by the TiO₂ the following day. The above effect was not the case for the pre-discharged composite metal oxides.

In this study, composite metal oxides derived from a sol–gel process of titanium and tungsten precursors (designated as WO₃@TiO₂ herein) were prepared and further used for mixing with a polymeric binder before coating onto metal substrates. The primary aim of this work was to investigate the effect of percentage loading of WO₃@TiO₂ on photocatalytic performance. The second objective was to compare the photocatalytic performance of the composite metal oxides (WO₃@TiO₂) with that of the mixed metal oxide (TiO₂/WO₃) prepared in accordance with optimal conditions described in our earlier report [19]. In this regard, it was hypothesized that more effective heterojunctions and greater photocatalytic performance of the composite metal oxides (WO₃@TiO₂) could be expected. Finally, the antifouling efficacy of polymer composite films containing WO₃@TiO₂ particles under a static field immersion test was of interest.

2. Results and Discussion

2.1. Characterization of Composite Metal Oxides

2.1.1. XPS Analysis

High-resolution XPS Ti and W spectra of the composite metal oxides (10% WO₃@TiO₂) are presented in Figure S5 in the Supplementary Materials. The presence of peaks at 458.83 and 464.53 eV can be noticed and these represent Ti 2p3/2 and 2p1/2 orbitals of TiO₂, respectively. The presence of these peaks indicates that the valence state of the Ti is 4+ [23]. Doublet splitting of the W 4f peaks at 35.2 and 37.1 eV was also noted. These represent the characteristic W 4f7/2 and W 5f5/2 signals, respectively, indicating the existence of W6+ state of WO₃ formed in the composite metal oxides [24]. The peak at 37.4 eV, overlapping with the W6+ 4f5/2 peak, was revealed after deconvolution, and that represents the Ti 3p orbital. In addition, the broad XPS peak centered around 38.6 eV in the W 4f spectrum of 1%WO₃@TiO₂ (Figure S5f was noted and can be assigned to the W 5p3/2 core energy level [25].

Figure 1 shows the XPS spectrum of O; 1s shows the presence of peaks at 529.6 and 530.6 eV, representing the oxygen bound to the Ti in the composite metal oxides. The shoulder peaks at higher binding energy (532.5 and 535.0 eV) were attributed to oxygen in the hydroxy groups and from the moisture on the surface of the composite metal oxides [17]. Similar XPS spectra were observed for the composites prepared using a lower percentage of tungsten precursor. However, the intensity of the peak that represents Ti 3p increased in the latter cases. This was due to the lower molar ratio of the W/Ti precursors used. From the XPS spectra of various composite metal oxides, the ratios of oxygen vacancies/lattices were also determined and summarized in

Table 1. Bandgap energy values, crystallite sizes, oxygen vacancies, surface areas, and rates of MB degradation (k) of the various metal oxides prepared via sol–gel process.

Sample Codes	Crystallite Size (nm)	Band Gap Energy (eV)	Oxygen Vacancy *	Surface Area (m2/g)	k (min−1)
TiO2 (sol–gel, rutile)	27.43	3.09	0.11	12.5245	0.01648
1% WO3@TiO2	28.75	3.27	0.07	78.2863	0.02460
3% WO3@TiO2	22.68	3.29	0.23	84.2878	0.03748
5% WO3@TiO2	26.66	3.29	0.27	28.3283	0.03755
10% WO3@TiO2	32.56	3.26	0.34	21.0851	n/a

* Referred to the vacancy/lattice ratios derived from the XPS spectra.

. The oxygen vacancies increased with WO₃ content, indicating that the presence of W element promotes formation of oxygen vacancies on the surface of TiO₂ due to the stronger W–O bond compared to Ti–O bond. This tendency favors formation of WO₃ with a perfect crystal structure during crystallization [26,27]. The presence of oxygen vacancies in the material significantly reduces its work function, leading to enhanced charge transportation at the interface [28].

2.1.2. Crystal Structure and Physical Properties

The XRD patterns of various composite metal oxides synthesized via the sol–gel process are illustrated in Figure S1. The XRD pattern of the neat TiO₂ shows the presence of peaks at 2θ = 27.4°, 36.07°, and 41.32°, representing the 110, 200, and 111 crystal planes of rutile phase TiO₂, respectively [29]. When the metal oxides were synthesized by introducing sodium tungsten dihydrate into the system, the XRD patterns show the presence of peaks at 2θ = 25.2°, 37.7°, and 47.9°, which can be ascribed to the 101, 004, and 200 crystal planes of the anatase phase TiO₂. A similar effect was also observed and discussed by Riboni et al. [17] and Joni et al. [30]. Specifically, the presence of a small amount of tungsten increased the thermal stability of the anatase phase TiO₂ and inhibited the transformation of the crystalline phase to rutile, which normally occurs at high temperatures [31,32]. Aside from this, the characteristic peaks representing monoclinic crystalline WO₃ cannot be seen in Figure 2a. This was probably due to the amorphous nature of the synthesized tungsten oxide obtained in this work [17]. The significant role of the amorphous/crystalline structure of the WO₃ in relation to its applications has been discussed in the literature. For example, Tatsuma et al. [19] pointed out that use of amorphous WO₃ is more favorable regarding photochromic applications. Cao et al. [33] also showed that the energy storage ability of TiO₂–WO₃ depends on several factors, including the molar ratios between the TiO₂ and WO₃, calcination temperature, and crystallinity of WO₃. Specifically, greater energy storage capacity can be obtained if the percentage crystallinity of the WO₃ is low. This is related to the looser structure of the metal oxide, which facilitates transportation of electrons and cations in the system. Additionally, the lack of a WO₃ peak might also be attributed to the concentrations of tungsten oxide in the composite, which were below the detection limit of the XRD technique. Tryba et al. [34] prepared photocatalysts from WO₃–TiO₂ and found that the presence of WO₃ could not be noticed when the percentage loading of the WO₃ was lower than 50. Li et al. [35] criticized that the lack of XRD peaks of the WO₃ implies that tungsten ions might be incorporated into the lattices of the TiO₂ by substituting the titanium ions via formation of W–O–Ti bonds. It was also possible that the tungsten ions resided in the empty space between the crystal lattices. Nevertheless, in our work, attempts were made to determine crystallite sizes of the various metal oxides using Scherrer’s equation. From Table 1, the crystallite sizes of the composites are in a range between 22.68 nm and 32.56 nm. Notably, the relationship between crystallite size and percentage tungsten precursor used was non-linear. This implies that insertion of W into the lattices of the TiO₂ is limited and formation of WO₃ domains could be more extensive and predominate. In this study, the maximum crystallite size (32.56 nm) was obtained when the maximum content of tungsten precursor was used (10%WO₃@TiO₂). The size is lower than that of the commercial TiO₂ particles (54.20 nm). In relation to photocatalytic performance, the lower the crystallite size of metal oxides, the greater the number of defects that can act as active sites for the photocatalytic reaction. However, care should be taken when optimizing crystallite size because excessive defects can induce more charge recombination and reduce the activities of the metal oxides [36].

Figure 2 shows UV/Vis spectra and the plots between the transferred Kubelka–Munk function (F(R)hν)^1/2 and the energy of the light absorbed of the various metal oxides. From the above plots, the bandgap energy values of the various metal oxides were determined and summarized in Table 1. The bandgap energy of the synthesized TiO₂ was found to be 3.09 eV, which is consistent with the typical value reported for rutile phase TiO₂ [37]. This value is also lower than that of commercial TiO₂ particles (3.40 eV), which contain an anatase structure. After coupling with tungsten, the cutoff wavelength of the synthesized metal oxides downshifted to a range between 377 nm and 380 nm. This corresponds to the bandgap energy values ranging from 3.26–3.29 eV. Previous work by Ismail et al. [16] found that the bandgap energy values of WO₃–TiO₂ decreased linearly from 2.93 eV to 2.74 eV as the content of WO₃ was increased from 0.5% to 5%. Owing to their similar ion radii, W⁶⁺, which also functioned as a dopant, can substitute TiO₂ (Ti⁴⁺) lattice and decrease the bandgap energy of the TiO₂. However, the above effect was not observed in this work. Herein, the bandgap energy values of the TiO₂ increased from 3.09 eV to the range of 3.26–3.29 eV after coupling with WO₃ due to the crystalline phase change discussed above.

2.1.3. Morphology

Figure 3 shows a typical SEM image, EDX spectrum, and EDX dot maps of the composite metal oxides (10% WO₃@TiO₂). The presence of Ti, W, and O elements on the surface of the composite metal oxides can be noticed. The EDX dot maps also show that the distribution of the W, Ti, and O domains are similar. The density of the W dots was low, however, because only 10% of the tungsten precursor was used for the synthesis. From the elemental analysis (see Section 3.2.1), the percentage mole of the W elements also increased with precursors’ feed ratios. However, some discrepancies between the two values (the W content in the feed and that in the products) were noticeable. This was attributed to the fact that the composition of the oxygen element in the products was not counted in the elemental analysis result. Additionally, the information gained from the SEM/EDX analysis represents the surface structure of the specimens and not that of the bulk sample.

Morphology of 10%WO₃@TiO₂ was also examined by transmission electron microscopy technique (TEM), as shown in Figure 4. Examination of the TEM image at high magnification revealed the presence of two distinct regions with different lattice fringes. These regions corresponded to the anatase TiO₂ (101) and the monoclinic WO₃ (130), with lattice fringes (d-values) of 0.352 nm and 0.22 nm, respectively [38,39]. At the interfacial region between the two phases, vague lattice fringes were observed (as indicated by the red dashed line area), indicating that development of a heterojunction structure at the interfacial region between TiO₂ and WO₃ via the sol–gel process [40]. Additionally, the lattice fringe observed on the TiO₂ particles lacked continuity (see yellow circle area in the inset of Figure 4), indicating the presence of defects that may have arisen from oxygen vacancies in the TiO₂ lattices, which was supported by the XPS results (Figure 1). However, these features were not observed in the TEM images of the mixed metal oxides (TiO₂/WO₃) (see Figure S2 in the Supplementary Materials), and the discrepancies can be attributed to differences in synthesis methods.

2.1.4. Photocatalytic Activities

Figure 5a shows the photocatalytic degradation of MB tested in suspensions containing different types of metal oxides. The activity of the mixed metal oxides (TiO₂/WO₃) was lower than that of the neat TiO₂ particles. This can be ascribed to a dilution effect in which the actual content of the TiO₂ decreased after mixing with WO₃. The performance of the mixed metal oxides increased slightly after introducing polythiophene (PTh) into the mixture (TiO₂/WO₃/PTh). The concentration of MB also kept decreasing after the UV lamp was off (after 100 min). This was not the case for the normal mixed metal oxides (TiO₂/WO₃ without polythiophene). The above effects can be ascribed to the presence of the conducting polymer (PTh) at the interface between the metal oxides domains, which facilitates transfer of electrons from the conduction band of the TiO₂ to that of the WO₃. Therefore, better photocatalytic efficacy of the mixed metal oxides can be expected [9,18].

Interestingly, when the composite metal oxides (WO₃@TiO₂) were used for the above testing, greater performance was noted. The concentration of MB decreased rapidly with UV irradiation time. The higher the percentage WO₃, the greater the catalytic activity.

Photocatalytic efficacy of composite metal oxides depends on various factors, including crystallite size, surface area, bandgap energy, and presence of defects. For instance, Munawar et al. [41] studied ZnO–Yb₂O₃–Pr₂O₃ nanocomposite and suggested that a small crystallite size and larger surface area of the composites contributed to greater formation of reactive oxygen species (ROS), which eventually led to destruction and death of the cell. Pandey et al. [42] demonstrated that antimicrobial activity of ZnO, ZnO–Ag₂O/Ag, ZnO–CuO, and ZnO–SnO₂ increased as bandgap energy of composites decreased. On the other hand, Widiarti et al. [43] reported that the band gap energy of CuO–ZnO composites increased with percentage concentration of CuO, and antimicrobial activity of the composites also increased with band gap energy values. In this study, we observed that photocatalytic activity of WO₃@TiO₂ did not show a linear correlation with surface area and crystallite size (refer to Table 1). Conversely, rate of MB degradation increased with the oxygen vacancy values of the composites. It appears that oxygen vacancies are the predominant factor influencing the efficacy of the photocatalysts. The oxygen vacancy formed near the interface between the composite metal oxides acted as h+ trapping species, which suppresses charges recombination.

The photocatalytic activity of the sol–gel-synthesized TiO₂ was also inferior to that of the composite metal oxides. This is because the crystalline structure of the synthesized product is rutile, which has a lower band gap energy value (Table 1). Additionally, charge transport resistance of the metal oxides was investigated by electrochemical impedance spectroscopy (EIS) technique. From the Nyquist plots (Figure 6a), it was found that the radius of semi-circle of 10% WO₃@TiO₂ was significantly smaller than that of the neat TiO₂. The charge transport resistance of 10% WO₃@TiO₂ and TiO₂ were 151.7 Ω and 82.6 Ω, respectively. This suggests that, by coupling TiO₂ with WO₃, recombination of the electrons and holes could be suppressed. This was related to the presence of the heterojunction in the composite metal oxides (WO₃@TiO₂), which facilitated transfers of electrons and holes from the CB of WO₃ to that of the TiO₂ and from the VB of the former to that of the latter, respectively [13].

To gain better understanding of the charge transport mechanism in the composite metal oxides (10%WO₃@TiO₂), an additional experiment was conducted by introducing various scavengers, such as silver nitrate (AgNO₃), potassium iodide (KI), p-benzoquinone (p−BQ), and isopropanol (IPA), into the system. The photodegradation of methylene blue was then measured after 120 min of UV light irradiation, and the photocatalytic activity in the dark condition was also tested after turning off the UV light for 30 min. More details on the experimental conditions and the results obtained can be found in the Supplementary Materials. From Figure 6b, it was found that the effect of KI, used as a hole scavenger, was negligible. On the other hand, the addition of AgNO₃ as an electron scavenger significantly suppressed the photodegradation efficiency of the metal oxides catalyst. This suggests that electrons are an important species in controlling photocatalytic performance of the metal oxides. Therefore, in this study, a Z-scheme-type charge transport mechanism was proposed as shown in Figure 7.

Furthermore, the rate constants of the various metal oxide catalysts were determined from the relationships between ln C/C₀ and UV irradiation time (Figure 5b). The rate constant of mixed metal oxides (TiO₂/WO₃) was 7.94 × 10⁻³ min⁻¹, which is much lower than that of the composite metal oxides (WO₃@TiO₂) (Table 1). The discrepancy can be ascribed to the different types and compositions of the raw materials used for preparing the samples. These factors led to different morphology and homogeneity of the two metal oxides systems. Figure 8 shows an SEM image of the mixed metal oxide in which some agglomerated particles, with dimensions of less than 1 μm, were noticeable. From the EDX dot maps: the distributions of the W element and Ti element are complementary. Additionally, the density of the X-ray dots representing the Ti in the mixed metal oxides was higher than that in the composite metal oxides (Figure 3). This was due to the greater ratio of WO₃ used for preparing the mixed metal oxides. From elemental analysis, the percentage mole of oxygen (O), titanium (Ti) and tungsten (W) on the surface of the prepared mixed metal oxides was 16.94, 3.39, and 0.79, respectively. This corresponds to the percentage W/Ti molar ratio of 0.23. This value is different from the feed ratio. It is also higher than that of the synthesis of composite metal oxides (see Section 3.2.1). In our opinion, the above discrepancies reflect that the mixed metal oxides are kinds of heterogenous mixtures containing different materials, including the WO₃ and TiO₂ domains and mixed metal oxides with polymerized thiophene as a binder between the particles (TiO₂/WO₃/PTh).

Table 2. Summarized photocatalytic performance and antifouling efficacy of various metal oxides from the literature.

Metal Oxides	Sample Preparation	Important Finding	Refs.
WO3–TiO2	Sol–gel synthesis	Photocatalytic performance of TiO2 coupled with 3% of WO3 was greater than that of the neat TiO2	Ismail et al., 2016, [16]
WO3–TiO2	Sol–gel synthesis using Na2WO4 and/or W(C2H5O)6	Greater photocatalytic performance obtained when the tungsten loading was 3% by mole	Riboni et al., 2013, [17]
WO3–TiO2		Anticorrosion efficacy (in the dark condition) of the composite metal oxide and the concept of electron pool were demonstrated.	Tatsuma et al., 2001, [18]
WO3–TiO2 mono- and bi-layer films	ITO glasses coated with WO3/TiO2 films	Photochromic behavior of the bilayer film and the concept of energy storage substance in the system were demonstrated.	Tatsuma et al., 2002, [19]
TiO2/WO3 mixed metal oxides	Mixing of TiO2 with WO3 particles via in situ polymerization of thiophene	Catalytic activity against MB of the mixed metal oxides in the dark condition was demonstrated. Better performance was found in the system containing PTh.	Jaritkaun et al., 2016, [20]
TiO2/PPy/(TiO2–V2O5) composite metal oxides	Synthesis of TiO2–V2O5 particles Mixing of TiO2–V2O5 with TiO2 via in situ polymerization of pyrrole	Catalytic activity against MB of the composite metal oxides in the dark condition was demonstrated.	Piewnuan, et al., 2014, [12]
TiO2 and Ag-doped TiO2	Sol–gel using Ti(OBu)4 and AgNO3 solution Acrylate-based reins as a binder	Antimicrobial efficacy of TiO2 (0.1% loading) described in terms of percentage survival for Micrococcus sp. was 19%. Greater efficacy of the doped TiO2 was noticed	Ruffolo et al., 2013 [44]
CuO@TiO2	CuSO4·5H2O + trisodium citrate + TiO2	Photocatalytic degradation of dye reached 90% with use of 10%CuO@TiO2, Antifouling coating on PES membranes.	Nguyen et al., 2022 [45]
CuxO/ZnO	ZnO nanorods by a hydrothermal CuxO were formed next by photoreduction	Effect of morphology on antibacterial activity of the composites was studied. The nanoweb sample exhibited the lowest bacterial survival ratio in the dark	Gandotra et al. 2021 [46]
(Garlic and chitosan) doped WO3–TiO2	Sol–gel synthesis of TiO2 in the presence of WO3 loaded	Garlic-loaded WO3–TiO2 exhibited good antibacterial activity against E. coli for all four concentrations (250, 500, 750, and 1000 g)	Dhanalekshmi et al., 2020 [47]
ZnO–CuO	Chemical coprecipitation technique.	ZnO–CuO nanocomposite had greater antibacterial activity than the neat ZnO. This was ascribed to the larger number of defects that could trap electrons, leading to high amount of reactive oxygen species	Jan et al., 2019 [48]
WO3@TiO2	TiO2/WO3 mixed via in situ polymerization of thiophene WO3@TiO2 via a sol–gel synthesis Vinyl binder loaded with WO3@TiO2	Greater photocatalytic performance of WO3@TiO2 composites, compared with that of the mixed TiO2/WO3, was confirmed. Antifouling efficacy of polymer coatings loaded with 10%WO3@TiO2 under the real environment was demonstrated.	This study

summarizes important details and the main findings obtained from the literature. The catalytic activity and antimicrobial efficacy of the TiO₂-based composite metal oxides are versatile and depend on their combination and the preparation methods. Despite the above intensive studies, however, the antifouling efficacy of polymer paints containing WO₃–TiO₂ tested under actual marine environmental conditions has not been reported. It is also noteworthy that the preparations of some composite metal oxides, such as TiO₂/PPy/(TiO₂–V₂O₅) [11] and TiO₂/PTh/WO₃ [20], involve multiple-step reactions. On the other hand, synthesis of composite metal oxides (WO₃@TiO₂) in this work is less complex (as described in the experimental part). Further, the photocatalytic performance of WO₃@TiO₂ in this work was better than that of the TiO₂/PTh/WO₃ photocatalyst, as shown in Figure 5b.

2.2. Structure Properties of Polymer/WO₃@TiO₂ Composite Films

2.2.1. Morphology

Due to the greater performance of the composite metal oxides derived from the sol–gel process, they were selected for further mixing with the polymer binder to prepare the antifouling paint. Figure 9 shows SEM images of the polymer composite films loaded with (10% w/v) of the metal oxides (10%WO₃@TiO₂) particles. Again, the presence of a brighter phase representing the metal oxides domains can be noticed. The existence of metal oxides was also confirmed by the EDX pattern, which showed the presence of X-ray peaks of Ti(k_α), W(L_α), and O(K_α). From the EDX dot maps, these metal oxides domains are evenly distributed in the polymer matrix phase. Large particles attributed to agglomeration of the metal oxides can also be noticed. Moreover, the SEM image reveals that the metal oxides particles are intimately in contact with the polymer matrix phase. This feature reflects good interfacial adhesion between the materials. This was attributed to polar interaction between chlorine atoms and acetate groups of the polymer and the hydroxy groups on the surface of the metal oxides.

2.2.2. Photocatalytic Activities

The photocatalytic activities of the polymer composite films loaded with different types of WO₃@TiO₂ particles are illustrated in Figure 10. When the sol–gel-synthesized TiO₂ was used, the concentration of MB gradually decreased with UV exposure time until the lamp was turned off (after 180 min). Superior catalytic efficacy was achieved when the 10%WO₃@TiO₂ composite was used. This was attributed to the fact that the crystalline structure of the synthesized TiO₂ is rutile, whereas that of the TiO₂ in the composite is anatase. Moreover, WO₃ has high acidity and great affinity for absorbing H₂O and ^•OH species. Therefore, formation of hydroxy radicals (^•OH) is enhanced and greater photocatalytic activity of the 10%WO₃@TiO₂ composite can be expected. Noteworthy: under dark conditions (when the UV lamp was turned off after 180 min), the concentration of MB still decreased. This was not the case for the polymer films loaded with other composite metal oxides (1%WO₃@TiO₂, 3%WO₃@TiO₂, and 5% WO₃@TiO₂). The photocatalytic performance of the polymer film loaded with 10%WO₃@TiO₂ was better than that loaded with other composite metal oxides. In our opinion, the discrepancy can be explained in light of the oxygen vacancy, which could have arisen due to many factors, including a reduction reaction and a mismatch between the TiO₂ and WO₃ lattices. From Table 1, we found that the oxygen vacancy ratios of the composite metal oxides are linearly increased with percentage of WO₃. The value of 10% WO₃@TiO₂ was the highest, which explains the superior performance of the composite compared with other analogues.

2.2.3. Antibacterial Efficacies

In this section, the antimicrobial performance of the polymer composite films was evaluated under laboratory conditions using collected seawater for incubation of the specimens. After 7 days, the samples were extracted and the results summarized in

Table 3. Number of colonies and kinetic constants of samples treated by seawater for 7 days.

Samples	Number of Colonies (CFU/mL)	Kinetic Constant (k) (day−1)
Control	1.3 × 104	0.28
Substrate coated with the neat binder	1.5 × 103	0.53
Substrate coated with polymer/metal oxides (10%WO3@TiO2) composite (10% w/v)	1.0 × 103	0.58

. The number of colonies of the system treated with the control sample was 1.3 × 10⁻⁴. The value dropped significantly to 1.5 × 10⁻³ when the substrate was coated with the neat polymer binder (without metal oxides). This effect could be attributed to the intrinsic hydrophobicity of the vinyl polymer binder, which led to a decrease in the surface energy of the coating [49]. Consequently, resistance toward biofouling adhesion of the coating increased. Furthermore, when the coating was loaded with the composite metal oxide (10%WO₃@TiO₂), the CFU value dropped to 1.0 × 10⁻³ and the kinetic constant values increased from 0.50 to 0.58 (day⁻¹). The above changes indicate that the number of bacteria decreased further due to the greater antimicrobial efficacy of the composite metal oxides. Zacarías et al. [50] studied the antibacterial performance of carbon-doped TiO₂, and kinetic constant values (k) ranging between 0.3 and 1.1 (day⁻¹) were reported. The k values were significantly dependent on the lighting conditions. For example, the constants determined under visible light and UV irradiation conditions (12.5% and 100% flux) were 0.3 day⁻¹, 0.3 day⁻¹, and 1.1 day⁻¹, respectively. In our opinion, it is possible that some colonies were killed under the UV irradiation used to activate the photocatalysts.

2.2.4. Static Immersion Field Test

In this study, attempts were made to investigate the antifouling and antimicrobial performance of coatings under a marine environment in a field test. Figure 11a shows the changes in appearance of the various substrates after immersion in seawater with time. After 3 months, some tube worm deposited on the surface of the control substrate became obvious. This was not the case for the substrate coated with the polymer films containing the composite metal oxides. The percentage fouling coverage on the substrates was calculated and presented in

Table 4. Percentage fouling coverage of the substrates coated with polymer film containing 10% w/v of the composite metal oxides (10%WO₃@TiO₂).

Testing Period (Days)	Percentage Fouling Coverage
	Control Substrate	Substrates Coated with Vinyl Polymer/Metal Oxides Composite
30	30.04	18.67
60	36.80	8.04
90	42.89	16.35

. The coverage values for the control system (the bare substrate) after immersion in seawater for 30, 60, and 90 days were 30.04%, 36.80%, and 42.89%, respectively. Lower coverage values, which were 18.67%, 8.04%, and 16.35%, accordingly, were obtained when the substrate was coated with the polymer film containing 10%WO₃@TiO₂. It is noteworthy that the relationship between percentage fouling coverage on the coated substrate and the testing periods was non-linear. This can be attributed to the fact that development of biofouling is influenced by several factors, including the characteristics of seawater, such as salinity, temperature, pH value, dissolved salts, and concentration of oxygen. These factors may vary over time and with changes in the season. Regarding this study, it is possible that the actual factors during day 60 were different from those in the previous 60 days.

After that, the substrates were collected from the field test and extracted for an antimicrobial test in the laboratory. Figure 11b shows the appearance of the various samples. The number of colonies and kinetic constants were calculated and summarized in

Table 5. Antibacterial performance of various substrates tested under seawater.

Testing Period	Control Substrate (Without Binder)	Coating with the Binder Containing 10% w/v of the Composite Metal Oxides (10%WO3@TiO2)
	CFU	CFU	Kinetic Constants	Percentage Reduction of Colonies
Day 1	<10	<10	n/a	n/a
Day 7	7.6 × 104 (±3.05)	4.3 × 103 (±0.1)	0.4106	94.34 (±0.13)
Day 15	62 × 103 (±4.58)	4.6 × 103 (±1.04)	0.1735	92.58 (±1.68)
Day 30	174 × 103 (±13.11)	12 × 103 (±0.66)	0.0884	92.94 (±0.38)

. For the control (the bare substrate), the number of colonies increased rapidly with immersion time and reached 1.7 × 10⁻⁵ (CFU/mL) after 30 days. By contrast, the CFU value of the substrate coated with the polymer film containing 10%WO₃@TiO₂ reached 4.3 × 10³ after 7 days. This is also higher than the value obtained from the same sample immersed in seawater under laboratory conditions (1.0 × 10³, Table 3). In our opinion, the discrepancies can be attributed to the fact that the actual environmental conditions in the field test were different from those in the laboratory. Overall, the above results indicate that the antibacterial efficacy of the substrate coated with the polymer composite films containing 10%WO₃@TiO₂ was greater than that of the control.

3. Materials and Methods

3.1. Chemicals

Titanium (IV) isopropoxide (98%) was purchased from Thermo Scientific Inc. Sodium tungstate dihydrate (98%) was supplied from Chem-Supply Ply. Ltd. Isopropyl alcohol (99.8%) was obtained from Carla Erba^TM. Hydrogen peroxide (30% (w/w) was obtained from Sigma-Aldrich. All chemicals were used as received. The vinyl-based polymeric binder used was SOLBIN^®C obtained from the Nissin Chemical Industrial Co. Ltd. According to the technical data sheet, it contains vinyl chloride (87%) and vinyl acetate (13%) repeating units. The degree of polymerization (DP_n) and number average molecular weight (M_n) of the vinyl (co)polymer are 420 and 3.1 × 10⁴ g/mol, respectively. The glass transition temperature (T_g) is 70 °C and the solution viscosity is 150 (mPa.s).

3.2. Preparation of Metal Oxides

3.2.1. Synthesis of Composite Metal Oxides (WO₃@TiO₂)

The composite metal oxides were synthesized via a sol–gel process using titanium isopropoxide and sodium tungstate dihydrate (Na₂WO₄-2H₂O) as the precursors. Experimentally, 5 mL of titanium isopropoxide was dropped into isopropyl alcohol (10 mL, 99.98%) in an ice bath. After stirring for 30 min, 50 mL of deionized water was added and the solution was stirred for further 30 min. Then, 25 of hydrogen peroxide (30%) was added and the reaction was allowed to proceed for 2 h. At this stage, change in color of the content in the reaction flask from colorless to yellowish (and orange eventually) was noticeable. Then, an aqueous solution of Na₂WO₄-2H₂O (54.4 mg in 5 mL of water) was added and the content was stirred vigorously for another 2 h. The product was then transferred into an oven and dried at 80 °C for 48 and at 120 °C for another 6 h, respectively. Finally, the dried product was grinned and calcined at 450 °C for 3 h. Moreover, composite metal oxides with a variety of molar ratio between the two precursors were also prepared and studied (

Table 6. Compositions of the various WO₃@TiO₂ composites.

Sample Codes	Precursors’ Feed Ratios *		Elemental Compositions (mole %) of the Products #
	Weight Ratios	Molar Ratios	W	Ti	W/Ti
1% WO3@TiO2	1.15/98.85	0.99/99.01	0.19	16.52	0.015/1
3% WO3@TiO2	3.47/96.53	2.91/97.09	0.35	15.32	0.023/1
5% WO3@TiO2	5.80/94.2	4.76/95.24	0.51	12.32	0.041/1
10% WO3@TiO2	11.6/88.4	9.10/90.90	1.42	21.76	0.065/1

* Na₂WO₄-2H₂O/Titanium (IV) isopropoxide; ^# from EDX analysis.

).

3.2.2. Preparation of Mixed Metal Oxides (TiO₂/WO₃)

For comparison purposes, mixed metal oxides containing TiO₂ and WO₃ particles were prepared and studied. Herein, the mole ratio between commercial grade TiO₂ particles and the WO₃ powder used for mixing was 1/2. Moreover, to enhance interfacial contact between the metal oxide domains, mixing was carried out concurrently with in situ polymerization of thiophene. More details concerning the mixing and polymerization conditions were well described in our earlier report [20]. Structures of the mixed metal oxides products prepared in this work were re-confirmed by XRD (Figure S1), TEM (Figure S2), FTIR (Figure S3), and UV/Vis spectroscopy (Figure S4) techniques (see the Supplementary Materials). Noteworthy, from the UV spectra, bandgap energy of TiO₂/WO₃ and TiO₂/PTh/WO₃, calculated by using the Kubelka–Munk model, was 2.55 eV.

3.3. Characterizations

Elements and the electronic state on the surface of the prepared metal oxides were examined by using X-ray photoelectron spectroscopy. The XPS data were collected using a Kratos Axis Ultra XPS operated with a monochromatic Al K_α X-ray source.

To determine the oxygen vacancy, O 1s XPS spectra of the metal oxides were deconvoluted. Two types of the oxygen species can be categorized, which are lattice oxygen and oxygen vacancy. The former is noticeable at the binding energy of 529.6 eV, whereas the latter can be observed at the binding energy of 530.6 eV [51]. In this regard, the ratio of oxygen vacancy/lattice oxygen (V_O) can be calculated by using the Equation (1).

$$V_O=\frac{A_{oxygen vacancy}}{A_{lattice oxygen}}$$

(1)

where A_{oxygen vacancy} and A_{lattice oxygen} are the areas of oxygen vacancy peak and lattice oxygen peaks derived by the deconvolution of O 1s XPS spectrum.

Brunauer–Emmett–Teller (BET) surface area analysis of the metal oxides was determined by measuring adsorption of nitrogen gas using a Micromeritics: 3FLEX apparatus. Morphologies of the composite metal oxides and polymer/metal oxides films were determined by using a field emission gun scanning electron microscope (FE-SEM; Nova NanoSEM 450). The specimens were coated with gold prior to SEM analysis to avoid a charging effect. The SEM experiment was operated under accelerating voltage of 15 kV. Both secondary electron and backscattering electron detectors were used to collect the signals and generate both topological and compositional contrast images, respectively. In addition, EDX detector was used to analyze the elements existing on the surface of the specimens.

Transmission electron microscopy (TEM) technique was also used to examine the morphology of the composite metal oxides using a transmission electron microscope (JEOL 2100Plus), which was operated at 200 kV in bright field mode. The TEM specimen was prepared by dispersing of the metal oxides in absolute ethanol before dropping onto a carbon-supported copper grid (200 mesh). The grid was then dried in the vacuum oven at 65 °C overnight before TEM experiment.

Crystal structures of the metal oxides were examined by using an X-ray diffractometer (XRD, Bruker AXS D8-Discover) in the 2θ range of 10–80° using Cu K_α (λ = 0.15406 nm) radiation (λ = 1.54178 Å). The accelerating voltage and the current used were 40 kV and 40 mA, respectively. From the XRD patterns, crystal size can also be calculated by using Scherrer’s equation (Equation (2))

D = 0.9λ/ßcosθ

(2)

where

• D is the crystal size (nm)
• λ is the X-ray wavelength (CuK_α = 0.15406 nm)
• ß is the line broadening at half the maximum intensity (radians)
• θ is the Bragg angle (°)

Bandgap energy values of the composite metal oxides were determined by using UV/Vis spectrophotometer (Shimadzu, Columbia, MD, USA, SolidSpec-UV3700) equipped with a Labsphere integrating sphere diffuse reflectance accessory and using BaSO₄ as the reference material. The reflectance spectra were scanned over the wavelength ranging between 200–800 nm. The spectra were then transformed into the Kubelka–Munk function F(R) to separate the extent of light absorption from scattering. The Kubelka–Munk function F(R) was then plotted against the energy of the absorbed light (E). The bandgap energy was then calculated using the Kubelka–Munk model (Equation (3))

F(R)E^1/2 = [(1 − R)² hν/2R]^1/2

(3)

Charge transport resistance of the prepared catalysts was investigated using electrochemical impedance spectroscopy (EIS) technique. A standard three-electrodes system was used to measure the charge transport performance of the catalyst. Ag/AgCl and Pt were used as the reference electrode and counter electrode, respectively. To prepare the working electrode, 0.1 g of catalyst powder was dispersed in 150 µL of absolute ethanol and the mixture was ultrasonicated in an iced bath for 30 min to homogenize the paste. The paste was coated onto a clean FTO conductive substrate using sticky tape to create a spacer, with the area of the spacer fixed at 1.5 × 2 cm² and the thickness of the coating controlled by the thickness of the tape. The catalyst-coated FTO substrate was then dried in a vacuum oven at 120 °C for 3 h before the test. The EIS experiment was performed over a frequency range of 100 kHz to 0.01 Hz at an amplitude of 10 mV. The working electrode was biased with a DC voltage of +0.7 (vs. Ag/AgCl, and the reaction was performed in 1 M NaOH solution. The impedance was measured using a VersaSTAT4 potentiostat (AMETEC, Princeton Applied Research, Oak Ridge, TN, USA) under dark condition.

3.4. Preparation of the Coating

About 0.1 g of the metal oxide particles were mixed with 1 mL of the vinyl-based polymeric binder (Solbin^® C, Seongnma, Republic of Korea) (80% v/v, in a thinner solvent). The ratio between metal oxides and the polymer was kept constant at 1/4 by volume. The mixture was stirred vigorously for 20 min before coating onto a steel plate (SS400 type, 25 × 25 mm²) using a bar coating machine ((RK model coater (H.J. UNKEL Ltd., Bangbon, Bangkok, Thailand). Thickness of the coating layer was controlled by adjusting the gap between a rod-bar and the substrate. The coating was dried at room temperature overnight. The final thickness after drying is approximately 50 μm.

3.5. Photocatalytic Activities Test

Photocatalytic activities of the metal oxides catalysts, both in the form of suspended particles and polymer composite films, were evaluated by measuring the degradation of methylene blue (MB) under UV irradiation using a Philips lamp (TUV T8 F17 1SL/25), which generates UV-C radiation. The distance between lamp and sample was 20 cm and the actual power of the lamp was 4.5 Watts. Typically, about 0.05 g of samples were immersed in of an aqueous suspension of MB (0.01 mM, 250 mL). The content in the reaction flask was kept in the dark for 24 h to enable establishment of adsorption–desorption equilibrium. Next, the UV lamp was turned on and concentration of MB was measured as a function of time. About 2 mL of the solution was collected at an interval of 30 min and the concentration of MB was determined from the absorbance at 664 nm using a UV/Vis spectrometer (Thermo Scientific, Waltham, MA, USA, Genesys-10S). Attempts were made to examine catalytic activities of the metal oxides in the dark condition. In this case, after exposure to UV lamp for 100 min, the light was turned off and change in concentration of MB with time was continually measured. Similar experimental procedures were used for testing with polymer/metal oxides composite films coated on steel plates, excepting that the dark condition was commenced after exposure to UV for 180 min.

3.6. Antifouling Efficacy Test

In this study, two different approaches were used for the antifouling test. In the first approach, the coated substrates were immersed in a test tube containing the seawater, which was collected from the Royal Navy Shipyard Department, Chonburi Province. Accordingly, the seawater contains a Gram-negative strain (Pseudomonas sihuiensis) and a Gram-positive strain (Exiguobacterium profungdum). The initial CFU/mL value was 1.4 × 10⁵. The seawater was also centrifuged at 4000 rpm before use. After immersion for 7 days, the substrates were collected and dipped into salt water (0.85%) for 3 s before centrifugal at 50 Hz for 1 min. The solution was then transferred into agar plates containing Zobell Marine agar (HIMEDIA, M384). The specimens were incubated at 37 °C for 48 h. Antifouling efficacies of the various coatings were then determined by measuring the number of colonies (Equation (4)).

CFU/mL = (Number of colonies × dilution factor)/(Volume of culture spread)

(4)

Percentage reduction of bacteria was determined by using the following equations

$$\mathrm{Reduction}(\%)=(\frac{{\mathrm{N}}_0-\mathrm{N}}{\mathrm{N}})\times 100$$

(5)

Log reduction = log (N₀/N)

(6)

where N₀ = Initial Colony forming unit, N = Colony forming unit at the end.

Bacteria decomposition rate was determined by the Equation (7)

N = N₀ exp (−kt)

(7)

where k = kinetic constant (day⁻¹).

Alternatively, the antifouling performance was studied under the actual marine environment in the field test. Practically, the steel plates coated with the polymer composite film were immersed in the seawater, at the Royal Thai Navy Shipyard Department, Chonburi Province for up to 3 months. The samples were collected as a function of time. The substrates were then collected. After centrifugal, the solution was dropped and spread onto agar plates before incubation. The number of colonies, decomposition of bacteria, and percentage reduction of colonies was determined by using the above equations. Changes in appearance and percentage fouling coverage of the substrate were also examined as a function of immersion time.

4. Conclusions

Composite metal oxides containing titanium dioxide coupled with tungsten trioxide (WO₃@TiO₂) were prepared via a sol–gel process using various molar ratios of precursors. The crystalline structure, morphology, elementals profiles, electronic state, and band gap energy values of the various composite metal oxides were characterized by XRD, SEM-EDX, XPS, and UV/Vis spectroscopy, respectively. Photocatalytic activity of the metal oxides was evaluated by measuring degradation of methylene blue under both UV irradiation and in the dark. It was found that the catalytic performance of the composite metal oxides was greater than that of the mixed metal oxides (TiO₂/WO₃) prepared by simply mixing TiO₂ and WO₃ particles with in situ polymerization of thiophene. The most effective antifouling coating was developed by feeding 10% of sodium tungstate into the sol–gel reaction of titanium precursors, and the obtained composite (10%WO₃@TiO₂) was then mixed with a vinyl-based polymeric binder prior to coating on steel substrates. From the static immersion field test, using seawater, the antifouling efficacy of the coating was demonstrated. In this study, the lowest CFU value was 1.2 × 10⁻⁴, obtained from the substrate coated with composite binder immersed in seawater for 30 days. However, the technology readiness level of the developed antifouling paint containing the composite metal oxides (WO₃@TiO₂) has yet to be upgraded. These include some issues, such as the adhesion strength between the coating and metal substrates, long-term durability of the coating, which can be determined by accelerated weathering test, and possible side effects of this coating on the marine ecosystem. These will be the aspects of our future work.