Water contamination caused by textile dyes and other industrial dyestuffs has become a worldwide concern (Lellis et al., 2019; Liang et al., 2018). The most commonly used dyes are classified as organic contaminants, with colored waste released into the water streams during dying process—a primary source of environmental damage (Shindhal et al., 2021). Furthermore, toxic byproducts generated in wastewater as a result of oxidation, hydrolysis, or other chemical reactions are considered hazardous to human health (Dougna et al., 2015). Over the past decade, TiO₂ has shown potential for the removal of organic pollutants such as methylene blue (MB), methyl orange (MO), and phenols under UV light and little visible light and air to produce hydroxyl that destroys organic dyes in photocatalytic process (Basavarajappa et al., 2020). However, the lack of visible region absorption, among other issues, is a bottleneck in this process. TiO₂ nanostructures have attracted great attention in numerous applications including ceramics, photocatalysts, photovoltaic cells, photoanodes, and sensors (Wang et al., 2009; Bahar et al., 2017).

TiO₂ is known to exist in nature in three various crystalline phases; brookite, anatase, and rutile (Allen et al., 2018; Hanaor and Sorrell, 2011). Unlike the other crystalline phases, the anatase form has a greater surface area, is more stable, and has higher activity (Ola and Maroto-Valer, 2015). The chemical, physical, and photocatalytic properties of TiO₂ are greatly influenced by the phase of the structure, and morphology is a crucial factor in assessing the material’s appropriateness for discrete applications (Andronic et al., 2011). TiO₂ generally has a broad bandgap equivalent to 3.2 eV along with a high surface area and is relatively non-toxic; however, it only absorbs 5% of solar radiation (energy), with the remainder being UV-Vis (Casino et al., 2014; Anderson and Bard, 1995). Furthermore, TiO₂ has a high rate of charge recombination of photogenerated electrons and holes. Scientists have used several techniques, including doping, to address the limitations of TiO₂ so as to make it a more efficient photocatalyst (Anderson and Bard, 1995).

The chemical and physical properties of metal oxide nanomaterials can be controlled by the process used to prepare them, which in turn influences the photocatalytic activity (Zhou, 2020). TiO₂ nanomaterials have been successfully synthesized utilizing a variety of methods, including hydrothermal synthesis, micro-emulsion, physical vapor deposition (PVD), chemical vapor deposition (CVD), and sol-gel (Zhou, 2020; Karami, 2010; Dubey, 2018; You et al., 2014). The Stӧber sol-gel procedure, which was developed for the nuclear sector in 1960s, is one of the most commonly used methods for preparing metal oxides due to its ease of preparation, high degree of homogeneity, low cost (Dubey, 2018; Mohamad Saad et al., 2015), and the possibility of forming small particles at room temperature (Bahar et al., 2017; Id, 2018; Zhai et al., 1990; Attia et al., 2002). It involves controlling the NPs’ surface chemistry, size, shape, and photo-stability to further enhance photocatalytic efficiency. The studies referenced reported the hydrolysis of tetraethyl orthosilicate (TEOS), Si (OC₂H₅), under acid condition to form SiO₂-based glass (Zhai et al., 1990; Alothman et al., 1990). However, this approach results in nanoparticle aggregation, which lowers the effective surface area of metal oxides.

The surfactant is one of the key factors that influence the structure-composition and photocatalytic characteristics of TiO₂. Surfactants are used in this process to reduce and improve agglomeration. Surface directing agent (SDA) acts as a template and decreases surface tension while enhancing crystal nucleation and development in the reaction (Huang et al., 2014). It also serves as an effective medium for reducing the internal radius of particles, which in turn reduces their size and effectively addresses the aggregation issue (Feinle et al., 2015). In addition, surfactant and acid are used to trigger repulsive force between the particles as a result of proton adsorption that positively charges uncharged precipitates (Galkina et al., 2011).

Feinle et al. (2015) investigated the effects of surfactant and acid on sol-gel-produced MgO aerogels, resulting in surfactant to aid particle dispersion. Andrade-Guela et al. (2018) reported the phase formation of TiO₂ upon the application of hydrochloric and acetic acids as (hydrolysis agents) catalysts. Few studies are evident that narrow down to the effect of surfactant on TiO₂-photocatalyst for fine-tuning morphological, optical, and photocatalytic properties.

This study investigates the structure-morphology and photocatalytic MB dye degradation performance of TiO₂ upon the addition non-ionic surfactant (Triton X100) prepared by a sol-gel process. The optical properties, morphology, and phase compositions were characterized using ultraviolet–visible (UV-Vis), photoluminescence (PL) spectroscopy, X-ray diffraction (XRD), scanning electron microscopy-energy dispersive spectroscopy (SEM-EDS), high-resolution transmission electron microscopy (TEM), Brunauer–Emmett–Teller (BET), Fourier transform infrared (FTIR), thermogravimetric analysis (TGA), and X-ray fluorescence spectroscopy (XRF) methods.

In Figure 2, all XRD patterns exhibit highly crystalline structures indexed at (101), (004) (200), (105), (211), (204), (116), (220), (215), and (224), corresponding to 2θ = 25.2°, 37.5°, 47.8°, 54.2°, 55°, 62.7°, 68.09°, 70.2°, 75°, and 82°, respectively. These results were assigned to tetragonal TiO₂ anatase and matched JCPD standards 01-083-5916 and 01-075-2546 for TiO₂ and TiO₂-sa, respectively. The XRD pattern showed no additional peaks, indicating that the as-synthesized TiO₂-based has no impurities. Both TiO₂ structures preferred anatase orientation, which might be due to the application of a weak acid such as acetic acid at room temperature and low calcination temperature (350 °C) which aligned with Wong et al. (2014). It is worth noting that only anatase TiO₂ was identified here, with no rutile phase; this could be attributed to the low concentration of oxygen vacancies caused by high oxygen concentration during particle growth, preventing changeover from anatase to rutile phases (Zhao et al., 2007). The crystallite size of the as-prepared TiO₂ nano-powders was calculated from XRD patterns. Using the Debye–Scherrer formula, the average crystallite size of the TiO₂ nano-powder produced was determined (Wu et al., 2008; Thiruvengadathan et al., 2013; Wang et al., 2015) Eq. 1: D = 0.9 λ / β ⁡ cos ⁡ θ (1)

The crystallite diameter is represented by D, the wavelength of Cu by λ, the Bragg’s angle is θ, the full width at half maximum (FWHM) of the greatest intense diffraction planes is represented by β, Scherer’s constant is 0.9, and the correction factor allows for particle shape (Mahshid et al., 2007). The average crystalline sizes were calculated as 38.26 and 40.01 nm; XRD data are presented in Table 1.

TGA/DTA (Figures 3A, B) was employed to evaluate the composition/decomposition and thermal stability of the as-prepared TiO₂ nano-powders. The rapid decay observed around 90 °C–180 °C was ascribed to the decomposition of the adsorbed water molecules, followed by the loss of solvent (isopropyl alcohol) (Fang et al., 2015; Alothman, 2016; Phattepur et al., 2019). The results aligned with Fang et al. (2015) and Kubiak et al. (2020). It was observed that TiO₂-sa had the highest weight loss (%), which was also attributed to the decay of Triton X-100. After 350 °C, thermograph curves displayed a plateau, perhaps due to the complete loss of solvent and Triton X-100; at this region, TiO₂ is thermal stable (Phattepur et al., 2019). Notably, TiO₂-sa showed late decomposition, which indicated that it resisted heat better than TiO₂ nano-powder. The further decay of TiO₂-sa after 850 °C might correspond to the beginning of the transformation of anatase to rutile, as described in Li J. et al. (2020) and Wetchakun and Phanichphant (2008). In addition, Figure 3B indicates the derivative of the TGA, which represents the temperature values at specific weight loss. For instance, the first 17.02% loss happened at 45 °C, followed by 8.35% at 240 °C for TiO₂. In contrast to TiO₂-sa, the highest weight loss of 48.78% occurred at 330 °C.

Figures 4A and B represent BET results for N₂ adsorption/desorption isotherms at 77K, BET specific surface area, pore volume and BJH pore size distribution of TiO₂ and TiO₂-sa. It was observed that the isotherms of as-synthesized TiO₂ nano-powders demonstrated a typical type-II isotherm as the International Union of Pure and Applied Chemistry (IUPAC) defines the H3 type. It is noteworthy that a hysteresis loop can be seen in the p/p^o ranges of 0.6–1.0 and 0.75–1.0 for TiO₂ and TiO₂-sa isotherms, respectively. The obtained results suggested the existence of interconnected mesoporous TiO₂ nano-powders, similar to that reported in literature [34]. Furthermore, upon the addition of Triton X-100, loop hysteresis shifted to high relative pressures (p/p^o), indicating pore volume transformation. Kubiak et al. (2020) compared BET and BJH parameters for Triton X-100-assisted TiO₂ and pluronic P123-assisted TiO₂ and found that pore volume and specific surface area were enhanced and the hysteresis loop shifted to higher pressure regions for assisted TiO₂. BET-specific surface area, average pore size, and pore volume are summarized in Table 2. The incorporation of a Triton X-100 contributed to the increase of specific surface area—150 cm² g^-1 in contrast to its counterpart at 90 cm² g^-1. In addition, pore size distribution (Figures 5A, B: insert) was correlated to the above results, where two prominent peaks centered at around 10–30 nm. The average pore size was 0.13 nm for TiO₂ and 0.58 nm for TiO₂-sa. Nonetheless, mesoporous volume increased from 0.20 cm³ g^-1 for TiO₂ to 0.37 cm³ g^-1 for TiO-sa. Surfactant was able to modify pore size, pore volume, and specific surface area.

Figures 5A–F display the SEM-EDS images of TiO² and TiO₂-sa. TiO₂ showed well-defined nanoparticle surface morphology. Aggregated spheres could be caused by the nature of the sol-gel process. EDS revealed the existence of Ti and O, which confirmed metal oxide formation by sol-gel process, with the latter results aligning with FTIR (Figure 5A). Minor impurities are observed, such as Au and Pb, which are ascribed to gold-coating during sample preparation for SEM analysis. Small Pb (0.1 wt%) impurity may be attributed to contamination. It was observed that the surface morphology of TiO₂-sa improved agglomeration and reduced particle size upon the addition of a Triton X-100 surfactant. The results agreed strongly with TEM images (Figure 5). Huang et al., (2014) improved nickel particle dispersion by using various surfactants, including Triton X-100.

Figures 6A–F denote TiO₂-based materials which were investigated by HRTEM analysis. TiO₂ particles showed uneven morphology due to the coalescing of particles comprising big particles, single particles, and particle clusters forming larger aggregates. Average particle size was 65–85 nm. It was observed that the incorporation of Triton X-100 reduced the size of particles as chemical dispersant agents from 85 nm to 65 nm. Furthermore, the SAED pattern of the surfactant-assisted TiO₂ nano-powder exhibited a defined pattern, indicating a polycrystalline (polymorphic rings) nature Phattepur et al. (2019).

Figure 7 illustrates the FTIR spectra of TiO₂ –based materials. Typical metal oxide peaks were observed. The spectra showed the stretching vibrations of υ (O–Ti–O) around 580 cm^-1 and ѵ(O–Ti–O) at 780 cm^-1, which indicated the characteristics of TiO₂ formation (Liu et al., 2012). Strong bending vibrations appearing at 1480.00 cm^-1 and 1500.00 cm^-1 were assigned to titanol υ (Ti–OH) and water molecules, respectively (Zhao et al., 2007). The consecutive peak around 1620 cm^-1 was due to the asymmetric vibration of υ (C=O) bond and υ(Ti–O–C) bond from the used precursor, solvent, and hydrolyzing agent (Liu et al., 2012). The vibration around 2600 cm^-1 was attributed to υ (=C–H) bond of solvent and surfactant (Liu et al., 2012; Ravishankar et al., 2020). The broad peak at 3300 cm^-1 was ascribed to the hydroxyl (-OH) group from adsorbed water molecules (Murashkevich et al., 2008; Sharma and Lee, 2020). It is noteworthy that the weak intensity for υ(Ti–O–Ti) reveal the lack of a rutile phase which aligned with XRD results (Figure 2). The υ(–OH) peak centering at around 3300 cm^-1 was relatively weaker for TiO₂-sa, which might be due to the surface modification of TiO₂. In addition, the reduction in the intensity of (OH) vibration bands indicates the reduced amount of H₂O of crystallization in TiO₂-sa.

All samples (TiO₂ and TiO₂-sa) including certified reference material (CRM) were run in duplicates, and the data was expressed into two decimals (Table 3). Using Horwitz to determine the closeness and differences of the values, the data indicated that both samples and CRMs were within acceptable limits. In addition, the elemental composition of TiO₂ is illustrated in Supplementary Figures S1, S2 and Table 3. It is evident that the composition of TiO₂ is 82%–88%. The results indicate the high purity of the as-synthesized TiO₂. The results for certified reference materials (CRM) —NIST 2711—exhibit 0.54%, which is within acceptable limits (0.54 ± 0.023) according to Horwitz, indicating the reproducibility, 99.99% recovery, and low relative standard deviation (RSD %). The later results show that the acceptable method developed using XRF Analyzer WingUI software is valid.

The UV-vis absorbance spectra of as-synthesized TiO₂ are illustrated in Figure 8A. TiO₂ and TiO₂-sa is dominated by the visible region around 475.00 and 500.00 nm, which is more redshifted than known TiO₂ absorption (380.00 nm). In addition, upon the addition of Triton X-100 on TiO₂ during sol-gel process, a significant bathochromic shift (higher wavelength from 475.00 to 500.00 nm) and enhanced absorbance were observed. Dubey (2018) reported a surfactant-free agglomerated TiO₂ synthesized sol-gel process; the wavelength obtained was <400 nm with the bandgap of 3.38 eV. Figure 7B confirms these results by displaying Tauc plot to estimate the optical bandgap as described Zhao et al. (2007). The band gap was reduced from 3.12 to 2.98 eV upon the incorporation of Triton X-100 as displayed in Figure 8B.

Figure 9 compares the photoluminescence (PL) of TiO₂ and TiO₂-sa controlled at the different excited wavelengths of 350 nm and 400 nm. It was established that the curves are similar in shape, suggesting that surfactant did not lead to a new PL phenomenon (Lellis et al., 2019; Murashkevich et al., 2008; Wang et al., 2018). However, TiO₂-sa exhibited blue-shift. The intense peak observed at around 440–480 nm is generally attributed to a hole–electron recombination [42]. At this wavelength, TiO₂-sa indicates the low intensity which may be ascribed to the reduction of the recombination reaction compared to the TiO₂ counterpart (Shindhal et al., 2021). The other peaks in a broad region from 481 to 561 nm originated from oxygen vacancies and Ti³⁺ (Zeng et al., 2019; Kasinathan et al., 2016), which is the transition of two electrons trapped in conduction to valance bands (band–band) (Dastan and Chaure, 2014). Generally, the increase in oxygen vacancies and surface defects corresponds to the increase in surface area (Wang et al., 2009; Chauhan et al., 2020). Moreover, this region (481–561) is attributed to the emission of bandgap transition and excitonic PL that resulted from the surface O₂ vacancies and surface defects. The PL of TiO₂-sa showed an enhanced surface defect induced by an enhanced surface area upon the addition of surfactant, suppressing the direct electron–photon (e−/h⁺) pair recombination (Li D. et al., 2020). These defects tend to create a new energy level just below the valence band which may create a trapping site and prevent recombination (Niu et al., 2020). This result indicates that TiO₂-sa exhibited higher surface area upon the addition of surfactant. It is believed that the surface area, crystallinity, and phase play a big role in the photocatalytic removal of organic pollutants such as MB, including textile dyes. Therefore, the PL results have a strong agreement with the photocatalytic results in Figure 10A.

Figures 10A and B denote the photodegradation curve of MB from 200 to 800 nm (Zhang et al., 2019; Hou et al., 2018; Wang et al., 2018). It was observed that absorbance and concentration decrease with time. The degradation efficiency (X) of the dye was calculated using Eq. 2 X % A 0 − A t A 0 − A t × 100 (2)

where X (%), A ₀ , and A _t represent the degradation efficiency, initial absorbance in zero irradiation, and absorbance after a given time interval, respectively. It can be seen that the dye photodegradation percentage increases with time until it reached 96.9% after an hour with TiO₂-sa (Figure 10A). Furthermore, TiO₂-sa showed an improvement in degradation as compared to its counterpart (TiO₂ = 75.36%) after 60 min of visible light irradiation. Due to this later result, it was believed that the photodegradation reaction rate followed the first-order Langmuir–Hinshelwood kinetics model. Therefore, the degradation rate was studied using Eq. 3: ln A 0 A t = k t (3)

where the degradation rate constant kt (k, min^-1) was determined from the slope of the straight line of l i n C 0 C t against the time interval t (Figure 10C) as a function of the experimental parameters used. The value was 0.03 and 0.05 for TiO₂-sa and TiO² respectively.

The mechanism of activity of TiO₂ photocatalysts in MB degradation is depicted in Figure 11. In short, the TiO₂ photocatalysts produce holes in the valence band when they are exposed to UV light because of an electron transfer from the valence to conduction bands (1) (Mapukata and Nyokong, 2020; Sakar et al., 2019). The electrons and holes thus formed can readily reduce and oxidize the pollutants adsorbed on their surface. This is due to their ability to facilitate the formation of superoxide radicals (O^2⋅−) from atmospheric oxygen (2) and hydroxyl radicals (^⋅OH) from water (3). The photo-generated radicals can, in turn, oxidize and degrade organic materials like MB into CO₂ and H₂O (Kim et al., 2011; Madkhali et al., 2023).

A facile sol-gel was successfully applied to synthesize anatase TiO₂ nano-powders at low preparation and calcination temperatures. It was found that the addition of Triton X-100 modified the structure, morphology, and optical properties of TiO₂-based materials. Characterization of TiO₂ revealed that the use of Triton X-100 surfactants played a vital role by reducing the average crystalline size and triggering the polycrystalline phase. TEM images showed reduced particle size when surfactant was introduced. TGA confirmed that calcination at above 350 °C completely removed the surfactant, showed thermal stable TiO₂ based nano-powders. In addition, the incorporation of surfactant enhanced surface area and optical properties through UV-vis red-shift up to the visible region (510 nm) and optical bandgap reduction. The PL of TiO₂-sa showed enhanced oxygen vacancies induced by an enhanced surface area upon the addition of surfactant which inhibited the direct electron-photon (e/h⁺) pair recombination. These results agree strongly with photocatalytic results. TiO₂-sa showed plausible structural and optical properties and higher photocatalytic dye degradation than its counterparts (pure TiO₂), making it a promising photocatalyst for the removal of organic pollutants in water.