Influence of Fe Ions on the Surface, Microstructural and Optical Properties of Solution Precursor Plasma-Sprayed TiO₂ Coatings

Abstract

This work investigates on how Fe incorporation influences the surface, microstructural, and optical properties of solution precursor plasma-sprayed TiO₂ coatings. The Fe-TiO₂ coatings were prepared using titanium isopropoxide and iron acetylacetonate as precursors, with ethanol as the solvent. X-ray diffraction analysis revealed the existence of both anatase and rutile TiO₂ phases, with a predominant rutile phase, also confirmed by Raman spectroscopy. There was an increase in the anatase crystals upon the addition of Fe ions. A longer spray distance further enhanced the anatase content and reduced the average TiO₂ crystallite sizes present in the Fe-added coatings. SEM cross-sectional images displayed finely grained, densely packed deposits in the Fe-added coatings. UV-Vis spectroscopy showed visible-light absorption by the Fe-TiO₂ coatings, with reduced band gap energies ranging from 2.846 ± 0.002 eV to 2.936 ± 0.003 eV. Photoluminescence analysis showed reduced emission intensity at 356 nm (3.48 eV) for the Fe-TiO₂ coatings. These findings confirm solution precursor plasma spray to be an effective method for developing Fe-TiO₂ coatings with potential application as visible-light-active photocatalysts.

1. Introduction

Hydrogen production through photocatalytic water splitting utilizes semiconductor materials like titanium dioxide (TiO₂) to absorb light energy and catalyze the splitting of water into hydrogen and oxygen, offering a promising avenue for sustainable and environmentally friendly hydrogen generation [1,2,3]. While TiO₂ is widely used in photocatalysis, its efficiency is limited by a high optical band gap (~3.2 eV) and a high recombination rate of electron–hole pairs (EHPs) [4,5]. To enhance its photocatalytic activity, it is necessary to extend the light absorption of TiO₂ to the visible spectrum, which can be achieved by modifying its optical properties through incorporation of other elements and compounds such as transition metals and metal oxides [6,7,8]. Metal/semiconductor (M/S) composites enhance photocatalytic performance by improving the charge separation at the semiconductor interface and increasing the visible light absorption [9]. Numerous metal-added TiO₂ nanocomposites with various structural variations have been developed for hydrogen production, such as Cu-TiO₂, Au-TiO₂, Pt-TiO₂, and others [10,11,12,13]. The introduction of foreign elements (dopants) into the surface of a material alters its electronic structure, which could result in increased light absorption, improved charge carrier separation, and enhanced overall efficiency. This approach makes doped coatings a promising approach for advancing the performance of photocatalysts.

The transition metal iron (Fe) is a suitable dopant for TiO₂ due to the similar ionic radii of Fe³⁺ (0.64 Å) and Ti⁴⁺ (0.68 Å) [14,15]. The addition of Fe narrows the band gap of TiO₂ by introducing intermediate energy levels, enabling visible light absorption and a photo-response [16,17]. Moreover, Fe³⁺ ions can act as electron–hole traps, reducing the recombination rates as the energy levels of Fe²⁺/Fe³⁺ align close with those of Ti³⁺/Ti⁴⁺ [14]. This improves the ability of TiO₂ to absorb visible light effectively. Several methods, including the sol-gel method [18,19,20], metal organic chemical vapor deposition [21,22], and reactive magnetron sputtering [18], have been used to deposit Fe-added TiO₂ coatings, with studies reporting enhanced photocatalytic activity of TiO₂ under visible-light irradiation, as reported elsewhere [19,23,24,25].

The group led by Dholam [18] demonstrated that Fe-doped TiO₂ coatings derived via the sol-gel method achieved high H₂ production rates during photocatalytic water splitting. Despite this result, this coating fabrication method is time consuming, which limits its scalability. Furthermore, it results in fine porosity [26], which limits its efficiency. High porosity is important in order to expose the holes to a high surface area in the smallest volume possible in order to react with water [27]. Another work also presented enhanced H₂ production using Fe(III)-doped TiO₂ nanoparticles compared to pure TiO₂ nanoparticles [28]. However, nanopowders pose health risks [29], so their usage should be limited.

The top layer of the metal-oxide-based hybrid photocatalyst plays a critical role in their overall functionality. They are ideal for adsorption due to their high surface area and microstructural features [30]. The active sites and high surface area enable influential adsorption events with excellent chemical stability [31,32]. During light excitation, processes such as electron-phonon scattering, irreversible electron transition, involvement of holes during oxidation, and electrons undergoing reduction reaction significantly influence charge recombination and, consequently, photocatalytic efficiency. Thus, understanding the processing technique and precursor composition and properties is essential for enhancing the performance of photocatalytic coatings.

The present work is focused on the investigation of the influence of Fe ions on the surface, microstructural, and optical properties of TiO₂ coatings developed using the solution precursor plasma spray (SPPS) process. This innovative method allows for the production of homogenous large-area coatings and strong substrate adhesion over conventional methods [33,34]. While TiO₂ coatings developed using SPPS have been studied [35,36,37,38,39,40], there are only a few reports regarding SPPS-deposited TiO₂ incorporated with metal ions and analysis of their properties for photocatalytic applications [40,41,42]. This study provides a rare insight into how Fe incorporation affects the light-responsive behavior of SPPS-deposited TiO₂, which remains largely unexplored in the prior literature.

2. Experimental Section

2.1. Solution Precursor Preparation

A 1.0 L 5 mol% Fe-added TiO₂ solution was used in the plasma spray process. The TiO₂ solution precursors were prepared by dissolving titanium isopropoxide (TTIP, Sigma-Aldrich, St. Louis, MO, USA, 97%) in an absolute ethanol (Scharlau, Barcelona, Spain, 99%), achieving a concentration of 0.6 M, followed by continuous stirring at 300 rpm using a magnetic stirrer. Simultaneously, the required amount of iron(II) acetylacetonate (Sigma-Aldrich, 97%) was thoroughly dissolved in ethanol by vigorous stirring at 500 rpm while heating to 60 °C. The TTIP and ethanol solution was then gradually diluted by adding the stirred Fe⁺–ethanol solution dropwise. To avoid gelation and ensure thorough mixing, a final stirring step of 30 min was performed.

2.2. Operational Plasma Spray Parameters

A direct current SG-100 plasma torch (Praxair S.T., Indianapolis, IN, USA) was used to deposit the TiO₂ and Fe-TiO₂ coatings on grit-blasted stainless-steel substrates with a thickness of 2.0 mm. Argon (Ar) and hydrogen (H₂) were used as the primary and secondary plasma working gases, respectively. The plasma spray experiments were conducted at stand-off distances of 40, 50, and 60 mm. The plasma torch was controlled at a speed of 500 mm/s to ensure successful deposition of individual particles on the temperature-controlled substrates. The direct deposition of TiO₂ and Fe-TiO₂ coatings via SPPS has been previously reported [43,44]. The detailed spray parameters are listed in

Table 1. Spray parameters for the solution precursor plasma spraying of TiO₂ and Fe-TiO₂ coatings.

Parameters	Values
Plasma power	28 kW
Ar flow rate	45 slpm
H2 flow rate	5 slpm
Plasma torch velocity	500 mm/s
Solution feed rate	35 g/min
Stand-off distance	40, 50, 60 mm
Substrate	Stainless steel
Nozzle diameter	0.2 mm
Injection pressure	1 bar
Number of spray passes	10 cycles

based on the reference work [44], while

Table 2. Sample labeling of the sprayed TiO₂ and Fe-TiO₂ coatings.

Stand-Off Distance	Coating Label
	TiO2	Fe-TiO2
40 mm	40TiO2	40Fe-TiO2
50 mm	50TiO2	50Fe-TiO2
60 mm	60TiO2	60Fe-TiO2

presents the samples along with the specific parameter values used during spraying.

2.3. Characterization of the Coatings

Scanning electron microscopy–energy dispersive X-ray (SEM-EDX) spectroscopy was used to examine the microstructure of the plasma-sprayed TiO₂ and Fe-TiO₂ coatings at low and high magnifications. Cross-sectional images of the coatings were obtained at 5000× magnification using a Tescan Vega3 SEM (Brno, Czech Republic) equipped with a backscattered electron (BSE) detector. Imaging was conducted under the following conditions: accelerating voltage of 20.0 kV, view field of 60.7 µm, and working distance of 14.19 mm. Moreover, the internal coating structure was analyzed on polished cross-sections at 80,000× magnification using SEM/Xe-PFIB FEI Helios G4 PFIB CXe microscope (Hillsboro, OR, USA) under a 2.00 kV accelerating voltage, utilizing a 2.59 µm view field and a 4.2 mm working distance. Additionally, elemental mapping analysis was conducted to assess the spatial distribution of the elements within the coating. The crystallinity and crystal structure of all the samples were characterized using X-ray diffraction (XRD) using an Ultima IV (Rigaku, Tokyo, Japan) system utilizing Cu Kα radiation with a wavelength of 1.54056 Å. The XRD patterns were recorded over a 2θ range of 10° to 70° with a scanning rate of 2°/min at room temperature. This was also used to determine the crystalline phase content and average crystallite sizes of all the SPPS coatings. The Raman scattering (RS) spectra were obtained using a LabRam HR800 (HORIBA/Jobin-Yvon, Kyoto, Japan) spectrometer fitted with an Ar⁺ laser at 514.55 nm. The laser power was kept at 50 mW while data was being collected within the spectral range from 50 to 4000 cm⁻¹. Photoluminescence (PL) spectroscopy was used to analyze the spontaneous light emission from a photo-excited material, providing insights into the molecular adsorption and surface reactions in photocatalytic material. The PL spectra were detected using a HORIBA FluoroMax Plus spectrofluorometer at room temperature with an excitation wavelength of 300 nm. The ultraviolet–visible (UV-Vis) light absorption of the coatings was evaluated on the basis of the reflectance spectra. The UV-Vis spectra were acquired using a QE6500 spectrometer (Ocean Optics, Orlando, FL, USA) with a coupled halogen–deuterium lamp as a light source in the wavelength range of 250–800 nm. The diffused reflectance spectra were used as another approach to estimate the band gap of the coatings. The Tauc plot ${\left(F\right(R)\times E)}^{1/2}$ was plotted as a function of the incident photon energy. The band gap of each coating was determined by fitting the reflectance spectra data according to the following equation:

$${\left(F\left(R\right)\times E\right)}^{1/2} = A\left(h\nu -E_g\right),$$

(1)

where $F\left(R\right)=K/S$ is the Kubelka–Munk function ($K$ and $S$ are the absorption and scattering coefficients, respectively), $h\nu$ is the photon energy, $A$ is a constant number, and $E_g$ is the band gap energy.

As a supporting analysis, the solution viscosity was measured prior to the spraying experiments. Additionally, under the experimental conditions described, white nanometric powders were obtained, dried, and subsequently analyzed using differential thermal analysis–thermogravimetric analysis (DTA-TGA) using STA Regulus 2500 (NETZSCH, Waldkraiburg, Germany) to assess their thermal behavior. This analysis was also used to determine the phase transition temperature of TiO₂. The powder samples were heated from room temperature to 1000 °C at a rate of 10 °C/min in a nitrogen atmosphere. XRD analysis was also conducted on powders heated at 450 °C.

3. Results

3.1. SEM-EDX

Figure 1 and Figure 2 depict typical polished cross-sections of the SPPS TiO₂ and Fe-TiO₂ coatings. Unlike conventional atmospheric plasma-sprayed coatings, which typically exhibit coarse splat boundaries, these features are absent in the SPPS TiO₂ and Fe-TiO₂ coatings obtained in the current work. Notably, no horizontal or vertical fissures are observed in the coatings. All the coatings can be compared to a two-zone microstructure, characterized by grainy particles and compact regions, as previously described in the reference paper [45].

The magnified cross-sectional images of TiO₂ and Fe-TiO₂ reveal a homogeneous microstructure with good adhesion to the stainless-steel substrate. The dense TiO₂ coating’s internal structure, shown in Figure 1b, consists of finely grained particles with clearly defined grain boundaries, considered to be one of the zones mentioned. On the other hand, the internal structure of the Fe-TiO₂ coatings, shown in Figure 2b, features significantly finer and more tightly packed granules compared to the pure TiO₂ coatings. This difference can be attributed to the changes in the thermal behavior and particle formation of TiO₂ and Fe-TiO₂ solution droplets when introduced into the plasma jet. Furthermore, the thickness of all the coatings ranges from 5 to 12 μm, showing the versatility of SPPS in depositing thin and thick coatings. The SEM images of the other TiO₂ and Fe-TiO₂ coatings sprayed at 50 mm and 60 mm are provided in the Supplementary Information.

Additionally, the appearance of lighter spots indicated by arrows within the TiO₂ coatings is described as one of the two zones, which is also noticeable in Figure 2a. These spots are likely associated with processed TiO₂ agglomerates. This observation is supported by the EDS images in Figure 3 and Figure 4, which indicate a significant presence of Ti and O elements in Figure 3c,d and Figure 4b. Figure 3 illustrates the EDS mapping of the 60TiO₂ coating’s cross-section. The BSE micrograph in Figure 3a is the region selected for the elemental analysis boxed in Figure 1a. The Ti mapping, shown in Figure 3c, reveals its strong and widespread presence in the coating, which confirms that Ti is the main constituent of the coating. Figure 3d displays O uniformly distributed and co-localized with Ti, which signifies the formation of a pure TiO₂, especially as the C in Figure 3e only surrounds that specific dense region where T and O dominate. Fe, shown in Figure 3f, is barely detected in this sample, supporting the suggestion that it is Fe-free, in contrast to Figure 4. The EDS images in Figure 4a present the elemental mapping of the 60Fe-TiO₂ coating’s cross-section. Figure 4a shows the selected EDS mapping area. The coating appears as the top, relatively denser, clearly distinct from the substrate. The Ti mapping in Figure 4b shows an intense detection, which also confirms its majority in the coating. Figure 4c displays the elemental distribution of Fe, showing a slightly lower intensity in the coating region, which suggests co-deposition with Ti, and its presence may have influenced the morphology of the Fe-TiO₂ coatings. The low visible Fe signal may be attributed to the low concentration of deposited Fe ions, which is likely below the detection limit of the equipment. Similar to the O mapping in Figure 4e, it is distributed throughout both the coating and the substrate, which limits the ability to confirm TiO₂ formation solely from this data and necessitates verification through other characterization techniques.

3.2. XRD Results

The X-ray diffractograms of all the coatings were analyzed using the ICSD patterns for the anatase and rutile phases of tetragonal TiO₂. Figure 5 displays the XRD patterns of the TiO₂ and Fe-TiO₂ coatings at various spray distances, revealing that the coatings are polycrystalline and consist of both rutile and anatase phases. This bi-phasic composition is beneficial for photocatalysis, as TiO₂ photocatalysts containing both phases often exhibit significantly higher catalytic activity than either phase alone [46].

The addition of Fe ions influenced the peak intensities of the stainless-steel substrates, with increased austenite (111) and (200) peaks observed at 2θ = 43° and 2θ = 54°, respectively. The XRD spectra of the Fe-TiO₂ coatings support the EDS mapping results, showing minimal detectable traces of Fe and that Fe³⁺ ions may have substituted Ti⁴⁺ ions at some TiO₂ lattice sites. The absence of additional peaks corresponding to other TiO₂- or Fe-related phases suggests that no secondary crystalline phases are present when the high-temperature SPPS process (>900 °C) is used, in contrast to reports in the literature [47] on coatings produced by lower-temperature processes (<900 °C).

The percentages of the rutile and anatase phases in the coatings sprayed at varied stand-off distances were calculated from the XRD data. The following equation was used to calculate the anatase phase content ($f_A$) of the as-sprayed coatings using the peak intensities of anatase (101) and rutile (110) TiO₂ denoted by ($I_A$) and ($I_A$), respectively [44]:

$$f_A={\displaystyle \frac{I_A}{I_A+26I_R}}$$

(2)

For the pure TiO₂ coatings, the anatase-to-rutile ratios were 11.3% anatase and 88.7% rutile for 40TiO₂, 21.6% anatase and 78.4% rutile for 50TiO₂, and 26.4% anatase and 73.6% rutile for the as-sprayed 60TiO₂ coating. The incorporation of Fe further increased the anatase phase content: 40Fe-TiO₂ showed 20.5% anatase and 79.5% rutile, 50Fe-TiO₂ had 32.4% anatase and 67.6% rutile, and 60Fe-TiO₂ exhibited 39.4% anatase and 60.6% rutile. As shown in Figure 6, Fe addition to the TiO₂ coatings consistently reduced the rutile content while enhancing the anatase phase. Likewise, increasing the stand-off distance led to a further decrease in the rutile content and a corresponding increase in the anatase content.

The Scherrer formula below was used to estimate the crystallite sizes from the measured width of the crystal planes’ diffraction curves [48]:

$$t_{hkl}={\displaystyle \frac{0.9\lambda }{\beta \mathit{cos}{\theta }_B}}$$

(3)

where $t_{hkl}$ is the crystallite size, λ = 0.15045 nm is the wavelength of X-ray radiation, β is the peak intensity full width at half-maximum (FWHM), and ${\theta }_B$ is the Bragg angle of reflection of a specific crystal plane (hkl). Figure 7 shows the calculated average crystallite sizes for anatase and rutile, all of which are below 20 nm. Without Fe, the rutile crystals have comparable sizes for all the stand-off distances, which are around 14 nm, whereas the anatase crystallite sizes exhibit a broader distribution. In contrast, the Fe-TiO₂ coatings display a more uniform anatase crystallite size, while the rutile crystallite sizes vary depending on the stand-off distance.

Figure 7 also shows that adding Fe to the solution affects the average growth of TiO₂ crystallites as the stand-off distance changes. For the pure TiO₂ coatings, the average crystallite size increases with an increasing stand-off distance. However, in the presence of Fe, the trend is reversed, with the average crystallite size decreasing as the stand-off distance increases.

3.3. Raman Spectroscopy

Raman spectroscopy provides detailed structural information about TiO₂ coatings, making it an essential tool for characterizing their crystalline properties and allowing extended analysis of their phase composition. Figure 8 presents the Raman spectra of the TiO₂ and Fe-TiO₂ coatings, revealing sharp and well-defined peaks indicative of high crystallinity coating material. All the TiO₂-based coatings exhibited the prominent characteristic Raman peaks of rutile, corresponding to the symmetries of B_1g, E_g, and A_1g, confirming the presence of the rutile TiO₂ phase [49,50]. The B_1g mode at 180 cm⁻¹ is associated with bending vibrations involving Ti and O atoms within the coatings. The peaks at 400 cm⁻¹ and 600 cm⁻¹ correspond to the E_g and A_1g stretching and bending vibration modes of the oxygen atoms in the TiO₂ lattice, respectively [51]. These findings support the XRD results, further confirming that rutile TiO₂ is more abundant than anatase TiO₂ in the coatings. In addition, a broad vibrational peak between 235 and 270 cm⁻¹, arising from multiple-phonon scattering processes, is distinctly observed in all the samples [49].

3.4. Photoluminescence

Figure 9 shows the PL spectra of the TiO₂ and Fe-TiO₂ coatings excited at 300 nm. All the samples exhibit broad emission bands within the 360–520 nm range, with two notable regions: emissions around 340–375 nm, attributed to direct band-to-band recombination [52,53], and broader emissions between 380 nm and 500 nm, centered around 439 nm, associated with oxygen vacancies and surface defect states [53,54,55,56].

The strong emission at 356 nm (3.48 eV) for TiO₂ corresponds to direct band-to-band recombination [52]. In contrast, the Fe-TiO₂ coatings exhibit significantly lower luminescence intensity in this wavelength range, indicating that Fe incorporation effectively suppresses direct e⁻-h⁺ recombination. This quenching effect is attributed to Fe ions acting as electron-trapping centers, thereby enhancing the charge separation. The shifting of this peak observed at 360–370 nm in the PL spectra of Fe-TiO₂ in Figure 9d–f may prove the latter, reflecting the energy state of excited electrons prior to recombination, which suggests intermediate energy levels introduced by Fe ions. Additionally, the emission peaks observed between 380 nm (3.26 eV) and 500 nm (2.48 eV) in the TiO₂ samples, such as the emissions of wavelengths 382–384 nm, 403–405 nm, 425 nm, 431–442 nm, 466–468 nm, and 489–492 nm, are attributed to surface defects and e^--h⁺ recombination at oxygen vacancy sites within the coating structure [53,54,55,56]. The presence of Fe ions increases the number of oxygen vacancies and surface defects, as evidenced by the broad emission bands spanning this region, as illustrated by the 400–401 nm, 438–439 nm, 468 nm, and 495–496 nm emissions wavelengths. These key spectral features provide qualitative insights into the impact of Fe addition on the recombination behavior of EHPs, ultimately supporting the role of Fe in improving the charge separation efficiency within the TiO₂ matrix.

3.5. UV-Vis Results

The absorption curves of the pure TiO₂ and Fe-TiO₂ coatings are shown in Figure 10. As shown in the spectra, the pure TiO₂ coatings exhibited a strong broad absorption in the UV region, attributed to the optical band gap of TiO₂. The onsets of the absorption edges for the pure TiO₂ coatings are at 375 nm, 391 nm, and 399 nm for increasing spray distances. The Fe-TiO₂ coatings exhibit UV absorption peaks but also feature an absorption tail extending into the visible region. The localized spectra indicate that the onset of absorption occurs at 433 nm, 427 nm, and 424 nm for the Fe-TiO₂ coatings sprayed at 40, 50, and 60 mm distances, respectively.

Using the following equation, $\lambda =1240/E_g$, the calculated band gaps of the TiO₂ coatings are determined to be 3.310 ± 0.326 eV, 3.171 ± 0.434 eV, and 3.108 ± 0.414 eV for increasing spray distances. On the other hand, the band gap energies of the 40Fe-TiO₂, 50Fe-TiO₂, and 60Fe-TiO₂ coatings are 2.864 ± 0.377 eV, 2.904 ± 0.314 eV, and 2.924 ± 0.284 eV, respectively.

The band gap energies were obtained directly from the plot with the use of the baseline approach [57,58,59,60,61,62]. The band gap energies of the Fe-TiO₂ coatings are 2.846 ± 0.002 eV, 2.920 ± 0.003 eV, and 2.936 ± 0.003 eV for the 40, 50, and 60 mm stand-off distances, respectively, which align well with the results presented in Figure 10. The fitted band gap energies from the absorption spectra and the Tauc calculation are summarized in

Table 3. Comparison of the calculated band gap energies from the absorption spectra and Tauc calculations.

Stand-Off Distance	TiO2		Fe-TiO2
	${\mathit{E}}_{\mathit{g}} = 1240/\mathit{\lambda }$	Tauc Plot	${\mathit{E}}_{\mathit{g}} = 1240/\mathit{\lambda }$	TAUC PLOT
40 mm	3.310 ± 0.326 eV	3.328 ± 0.002 eV	2.864 ± 0.377 eV	2.846 ± 0.002 eV
50 mm	3.171 ± 0.434 eV	3.168 ± 0.007 eV	2.904 ± 0.314 eV	2.920 ± 0.003 eV
60 mm	3.108 ± 0.414 eV	3.139 ± 0.001 eV	2.924 ± 0.284 eV	2.936 ± 0.003 eV

. It was observed that the band gap energies of pure TiO₂ decreased with an increasing spray distance, while at the same time, the band gap energies increased for the Fe-TiO₂ coatings. This result can be related to the average TiO₂ crystallite size deposited on the substrate with and without the addition of Fe. As the spray distance increases, the average crystallite size of TiO₂ increases, while for the Fe-TiO₂ coatings, the crystallite size decreases.

4. Discussion

4.1. Influence of Fe Ions on the Microstructural Properties of TiO₂ Coatings

Addressing the grainy morphology of the coatings involves identifying the specific cause, which may be linked to the formulation, application process, or environmental conditions. A difference in the morphology between TiO₂ and Fe-TiO₂ coatings is observed. As illustrated in Figure 1, the Fe-TiO₂ coatings exhibit a fine-grained structure compared to the coarser morphology of the TiO₂ coatings. Since both coatings were obtained using the same spray parameters, the variation in morphology is likely due to the addition of Fe ions to the feedstock solution, which influenced the plasma environment during the spraying process.

The modification of the properties of the plasma vicinity altered the chemical and physical changes that the solution underwent. As reported in [63], there was an increase in the plasma ability of heating, which is caused by the Fe ions as additional mobile species present in the solution. This modification allows Fe-TiO₂ particle droplets to remain within the plasma for a longer duration, exposing them to higher temperatures than the TiO₂ droplets. As a result, the Fe-TiO₂ coatings consist of smaller deposited grainy particles. Additionally, this phenomenon can be inferred from the recorded substrate temperature during the spray process. Using thermocouples, the substrate temperature was measured, revealing a maximum of ~340 °C for TiO₂ and ~525 °C for Fe-TiO₂. The temperature of the substrate for TiO₂ remained relatively stable, whereas the Fe-TiO₂ substrate temperature increased progressively throughout the deposition process. A higher substrate temperature correlated with a higher plasma jet temperature, further explaining the formation of finer particles due to the rapid vaporization and fragmentation of the solution droplets during the spray process. Despite the morphological differences, the coating thicknesses for each spray distance remain comparable, as both coatings were deposited under the same spray conditions.

This claim is further supported by the viscosity measurements of the TiO₂ and Fe-TiO₂ solution precursors. Two flow curves were measured for each sample before the spray process, covering a shear rate range from 10 to 130 s⁻¹. The viscosity values were initially taken using an increasing shear rate ramp, followed by a subsequent decreasing ramp from 130 to 10 s⁻¹. Figure 11 presents the viscosity measurements of the prepared solutions. Both solutions showed similar average viscosity measurements for both ramp tests. However, a slight difference was observed, with the Fe-TiO₂ solution displaying a lower viscosity compared to pure TiO₂. Since the Fe-TiO₂ solution has lower viscosity than TiO₂, it has a greater tendency to form tiny droplets under primary atomization and evaporate during spraying [64]. This leads to the faster breaking up of droplets and the formation of fine molten deposits. Indeed, the size of the deposited particles is dependent on the viscosity and the evaporation dynamics of the precursor solution.

Additionally, the XRD results confirmed the presence of both rutile and anatase phases, with rutile being the dominant phase. This is advantageous for photocatalysis, as mixed TiO₂ phases generally exhibit higher catalytic activity than either of the component phases [46,65,66]. Both coatings displayed a more pronounced rutile (110) peak compared to the anatase (100) peak, indicating a higher fraction of rutile TiO₂ in the sprayed coatings. However, the phase percentage calculations at different stand-off distances revealed an increase in the anatase content and a drop in the rutile content with the addition of Fe ions. These observations can be explained by the thermal behavior of the powders used as coating precursors.

In Figure 12, the precursors undergo sequential processes of solvent vaporization, decomposition, crystallization, and phase transformation. The crystallization of amorphous TiO₂ into anatase TiO₂ is marked by an exothermic peak at a temperature of ~423 °C, as verified in its first derivative plot, and is comparable with the temperatures previously reported [34,67]. When Fe was added, the crystallization of TiO₂ shifted to a higher temperature of ~454 °C, indicating that Fe ions significantly influence the crystallization of TiO₂. The presence of Fe prolonged the formation of the anatase phase, probably due to Fe ions slowing down the growth of crystalline TiO₂ [68]. Despite this, the anatase-to–rutile phase transition temperature for TiO₂ and Fe-TiO₂ was at 812–824 °C. This also explains the increased anatase content when Fe was added, as a lower percentage of anatase was fully transformed into rutile TiO₂. In many cases, an increased anatase content tends to result in better TiO₂ photocatalytic activity, and its combination with rutile further enhances the catalytic activity [66,69,70]. From the literature, the synergistic interaction between the anatase (011) and rutile (110) planes enhances the charge separation and suppresses the electron–hole recombination, resulting in good photocatalytic activity [71].

Moreover, no characteristic peaks corresponding to iron oxide phases were observed in the XRD spectra of any of the coating samples. This can be attributed to the low concentration of Fe ions, and hence, could not be detected by XRD. This is further supported by the XRD spectra of the heated Fe-TiO₂ precursor powders shown in Figure 13, which also did not exhibit any diffraction peaks for Fe. This implies that Fe is below the detection limit and that the Fe³⁺ ions may have substituted Ti⁴⁺ in some of the TiO₂ lattice sites [72,73]. The incorporation of Fe ions into the TiO₂ lattice structure likely contributed to the prolonged TiO₂ crystallization temperature, thereby influencing the anatase and rutile phase content of the Fe-TiO₂ coatings. Moreover, the anatase peaks in Figure 13 are dominant upon the addition of Fe, which conveys an increased anatase content in the sprayed Fe-TiO₂ coating.

For both coating types, increasing the stand-off distance resulted in a higher anatase phase content, as particles experienced longer dwell times in the plasma jet. Conversely, a shorter stand-off distance caused a faster temperature rise, exceeding 600 °C, which promoted the transformation of nanocrystalline anatase into rutile TiO₂. This explains why the coatings sprayed at 40 mm developed larger rutile crystals, due to the increased heat exposure and anatase’s high surface energy, and also, the higher percentage of rutile formation. Thus, the farther the coating from the solution injection point, the fewer the anatase content that transforms into rutile TiO₂.

Regarding the crystallite size, the rutile phase exhibited uniform crystal sizes across all the stand-off distances in the TiO₂ coatings. However, the anatase crystals in the Fe-TiO₂ coatings maintained a consistent average size at all the stand-off distances. This implies that the incorporation of Fe atoms into some TiO₂ crystals effectively controlled their growth, resulting in a wider range of anatase crystal sizes.

4.2. Influence of Fe Ions on the Optical Properties of TiO₂ Coatings

All the TiO₂-based coatings exhibited the characteristic stretching peaks of rutile without any trace of anatase TiO₂. Since Raman spectroscopy is a point-to-point characterization technique, it is likely that rutile TiO₂ crystals were primarily detected during the measurement. This confirms the XRD results, which indicate higher rutile TiO₂ content than anatase in the coatings. Furthermore, there are no significant changes in the Raman spectra upon the addition of Fe to TiO₂, suggesting that the phase content and average crystal sizes of the TiO₂ and Fe-TiO₂ coatings are comparable. It is noteworthy to mention the appearance of multi-phonon scattering in the Raman spectra. This happens especially when the coatings are being excited by a laser pulse, generating an EHP, suggesting an electron upconversion via intervalley transition, particularly in nanostructured TiO₂ [74]. The excited electron, after relaxation, is prohibited from recombining in a hole that occupies a different valley [75]. As soon as the electron is delivered by upconversion, it can immediately recombine with the hole in the valence band, releasing a photon that the PL intensity detector can measure.

PL reflects charge carrier recombination near the semiconductor surface, making its intensity and spectrum sensitive to surface reactions and indicative of photocatalytic efficiency. Evidently, in Figure 9, recombination of the charge carriers is evident in both the TiO₂ and Fe-TiO₂ coatings. The differences in the PL spectra between these samples indicate that the Fe ion introduces new light-emitting phenomena. These observations are summarized in the proposed PL mechanism for Fe-TiO₂ coatings illustrated in Figure 14.

The process starts with electron excitation from the valence band to the conduction band and shallow states [76]. The relaxed photogenerated electrons in the conduction band can then follow multiple recombination paths [77]. In (path a), a direct recombination happened, emitting 3.48 eV of energy (356 nm), as observed in the TiO₂ coatings. The reduced PL intensity at this energy when Fe ions were introduced is attributed to the enhanced metal–metal (Fe-Fe) interactions, acting as luminescence quenchers [78].

Moreover, shallow states, which are initially empty electron traps, can be populated from band-tail states. These electrons can subsequently relax by settling into trap states, caused by Fe ions, located significantly below the conduction band edge. These trapped electrons will recombine to a hole in the valence band (path b) [79,80], like the PL emission with wavelength 361–370 nm (3.43–3.35 eV). Another dominant pathway for above-band gap excitation involves the recombination of conduction band electrons with deep empty states created by OH⁻ or trapped holes (path c) [76,81]. This can be presented by 468 nm light emission of 2.65 eV energy. Also, defects can cause PL emissions [82]. Several PL emission peaks, such as 400 nm (3.08 eV) and 439 nm (2.82 eV), are attributed to the presence of various defect states and impurities in the coatings. Defect emissions are the results of oxygen vacancies. These oxygen vacancies serve as recombination and/or carrier trap sites, offering alternative pathways for energy transfer and contributing to the additional emission features in the PL spectrum [79]. The emission at 2.52 eV results from the recombination of an e^- and an h⁺, where the electron is situated from an oxygen vacancy located in the sub-bandgap as a dense trap state (path d).

The presence of oxygen vacancies confirms the dominance of the TiO₂ rutile phase in all the coatings. Aside from the very high temperature of the plasma jet that helped attain the phase transformation, oxygen vacancies offer space for atomic arrangement and promote anatase–rutile transformation [83]. This also accounts for the higher PL peak intensities observed in the Fe-TiO₂ coatings compared to pure TiO₂. The substitution of Fe³⁺ ions into the TiO₂ lattice introduces oxygen vacancies as a result of charge compensation, which increases the PL emission [84]. It can be supported by the possible emergence of impurity energy levels within the band gap, primarily associated with Fe states. The Fe ions act as trapping sites that inhibit direct recombination. The PL emission energy of 3.48 eV, corresponding to the direct recombination of e^- and h⁺, aligns with the band gap transition energy. This is comparable to the calculated band gap energy of 3.33 eV from the UV-Vis spectroscopy via the Tauc plot.

The addition of Fe to catalyst-free TiO₂ precursors produces denser coatings composed of ultrafine-grained particles, which is beneficial for photocatalytic applications. Nanostructured coatings have a high surface atom fraction, improving the light absorption. In pure TiO₂, light absorption primarily occurs via interband electron transitions, which are direct but limited due to crystal symmetry constraints. Minimal absorption results, as direct transitions are typically disallowed. However, when absorption occurs at the crystal boundaries, such as surfaces or interfaces, it allows indirect electron transitions. These transitions significantly enhance the light absorption, which will contribute to improving the photocatalytic efficiency in Fe-TiO₂ coatings [85].

This means that when a material is made up of nanocrystals, just like the SPPS Fe-TiO₂ coatings, a reasonable enhancement of the absorption can be observed. As shown in Figure 10, the SPPS Fe-TiO₂ coatings exhibit significantly enhanced light absorption in the visible region. This enhancement can be attributed to the nanoscale dimensions of the particles, which result in a high surface-to-volume ratio and a substantial fraction of surface atoms. Notably, the increase in interfacial absorption becomes particularly pronounced when the particle size approaches ~20 nm or smaller [85,86]. The XRD results revealed that the TiO₂ crystalline sizes are less than 20 nm, which justifies the absorption of Fe-TiO₂ in the visible region.

The Fe-TiO₂ coatings have lower calculated band gap values than TiO₂. This confirms the successful incorporation of Fe into TiO₂ because of the potential appearance of certain impurity energy levels within the band gap that are primarily made of Fe. The valence and conduction bands gradually broaden in the presence of Fe, which causes the band gap to decrease. Thus, the addition of Fe ions greatly improves the activity of TiO₂ in absorbing visible light.

Plasma spraying of a catalyst-free Fe-TiO₂ solution precursors proves to be an effective method for producing nanostructured, crystalline Fe-TiO₂ coatings with excellent deposition quality and precisely controlled anatase content. Furthermore, the emergence of traps because of the shallow states caused by Fe ions and oxygen vacancies, the multi-phonon scattering, and the large surface-to-volume ratio of the SPPS Fe-TiO₂ coating creates the possibility of well-timed utilization of photogenerated carriers in interfacial processes for photocatalysis.

5. Conclusions

This study presents the development of Fe-TiO₂ coatings via the innovative SPPS process and explores the influence of Fe ion incorporation on the surface, microstructural, and optical properties of TiO₂. The properties of the TiO₂ coatings for photocatalytic application were found to improve upon the addition of Fe ions. Cross-sectional images revealed dense deposits with fine grains in the Fe-TiO₂ coatings, indicating the successful incorporation of Fe ions into the TiO₂ structure through substitution, facilitated by the close radii of Fe³⁺ and Ti⁴⁺. Structural analysis revealed the presence of both anatase and rutile TiO₂ phases, with the increased rutile content confirmed by both XRD and Raman spectroscopy. Interestingly, Fe incorporation also promoted anatase crystal formation, especially at longer spray distances, which additionally contributed to a reduction in the average crystallite size. These structural changes were accompanied by a significant enhancement in visible-light absorption and a decrease in band gap energy, from 2.936 ± 0.003 eV to 2.846 ± 0.002 eV, indicating improved photocatalytic potential under visible light. PL analysis revealed a decrease in the direct recombination of EHPs, as indicated by the reduced emission intensity at 356 nm (3.48 eV) in the Fe-TiO₂ coatings, along with the appearance of radiative emission at 361 nm, suggesting enhanced charge separation prior to recombination. Overall, the results highlight SPPS as a viable technique for depositing Fe-TiO₂ coatings with enhanced physico-chemical and optical properties, making them promising candidates for photocatalytic applications.