Structure and Properties of Silver-Platinum-Titanium Dioxide Nanocomposite Coating

Abstract

The aim of this study was to produce a coating for protective glass glued to touch displays with high antibacterial effectiveness. This paper presents the structural, mechanical, optical, and antibacterial properties of a TiO₂:Ag–Pt coating prepared by dual reactive DC and RF magnetron sputtering. Characterization techniques used include XRD, TEM with EDS, SEM, AFM, nanoindentation for hardness and Young’s modulus, wettability tests, and optical property analysis. The coating exhibited columnar crystals with a width of 30–50 nm. Crystals of anatase, rutile, silver, and platinum with a size of up to 3 nm were identified. The coating deposited on glass had a concentration of 5.0 ± 0.2% at. Ag and 4.4 ± 0.1% at. Pt. The value of the optical band gap energy, corresponding to the direct transition, was 3.36 eV, while Urbach’s energy was in the order of 500 meV. The hydrophilic coating had a roughness RMS = 1.8 ± 0.2 nm, hardness HV = 6.8 ± 0.5 GPa, and Young’s modulus E = 116 ± 8 GPa. A unique combination of the phase composition of the TiO₂:Ag–Pt coating, metallic Ag and Pt nanoparticles in a ceramic matrix of anatase and rutile crystallites resulted a >90% reduction of Staphylococcus aureus bacteria. This antibacterial effect was attributed to the activation of the doped semiconductor under visible light via plasmon resonance of the Ag and Pt nanoparticles, as well as a light-independent antibacterial action due to Ag⁺ ion release. In contrast, commercial antibacterial coatings typically achieve only around 60% bacterial reduction.

1. Introduction

As a popular photocatalyst, titanium dioxide (TiO₂) is widely used due to its availability, easy manufacturing and handling, non-toxicity, high thermal and chemical stability, low-cost, high refractive index, and, of course, high photocatalytic efficiency [1,2]. Photocatalytic TiO₂ is utilized across various industries. Its properties make it suitable for use in water treatment [3], antibacterial coatings or so called self-cleaning surfaces, and antifogging materials [1,4]. The number of publications dedicated to TiO₂-based photocatalysts increases annually [2]. TiO₂ photocatalysts can be obtained using different methods, such as chemical vapor deposition (CVD), pulsed laser deposition (PLD), atomic layer deposition (ALD), electron beam evaporation, liquid-phase deposition, sol-gel processes, and magnetron sputtering (MS) [5,6]. However, many methods of thin film deposition do not provide equilibrium conditions of growth. This is the case with magnetron reactive sputtering [7]. However, magnetron sputtering can be used to obtain a homogenous, high-density, nanometer-thick, hard coating with good adhesion to a substrate [8]. Moreover, the radio frequency (RF) magnetron sputtering method offers significant advantages in terms of controlling the structure and composition of the TiO₂ films, as well as in its adaptability to varying deposition conditions [5]. Reactive magnetron sputtering, as one of the PVD methods, enables the fabrication of nanocomposites composed of metallic silver (Ag) nanoparticles with sizes of several nanometers embedded within a ceramic TiO₂ matrix. Such structures provide increased antibacterial properties [9]. One effective approach to improving the photocatalyst performance is the incorporation of noble metals such as Pt, Au, Ag, and Pd, which enhance the charge separation efficiency [10]. The addition of a noble metal to a photocatalytic semiconductor can change its surface properties [4]. The doping of TiO₂ with plasmonic noble metals overcomes various limitations associated with the use of bare TiO₂, e.g., it lowers the band gap energy and allows the absorption of photons to occur at lower energy, i.e., toward the visible region in the electromagnetic spectrum, primarily due to the localized surface plasmon effect (LSPR). It also increases the lifetime of electron and hole pairs, which increases the photocatalytic efficiency of noble metal doped TiO₂ [11,12]. In addition to enhancing the photocatalytic activity, such dopants can also inhibit undesirable crystallite growth and phase transformations [6]. Among the most commonly used metals to improve the photocatalytic performance of TiO₂ in the visible light range are Pt and Ag. Metallic Pt nanoparticles act as electron trap centers, helping to prevent electron-hole pair recombination. Pt can also influence the crystal structure of TiO₂, leading to a partial phase transformation from anatase to rutile, which can beneficially affect the photocatalytic properties of the coatings. The photocatalytic properties of titanium dioxide doped with Pt and Ag prepared by the sol-gel process have been widely discussed in the literature. A. Zielińska-Jurek et al. [13] analyzed the inhibition of microorganism growth and photocatalytic activity under visible light irradiation for phenol degradation on Ag-Pt/TiO₂ photocatalysts. They showed that the photoactivity of Pt/TiO₂ and Ag-Pt/TiO₂ powders depends mainly on the Pt nanoparticle size; the smaller the size, the higher the efficiency of phenol degradation. DRS, XRD, TEM, and BET measurements confirmed that the most active photocatalyst under visible light (Ag-Pt/TiO₂) contains fine Pt NPs (approx. 3 nm) and small Ag nanoparticles (approx. 6 nm) deposited on a self-synthesized TiO₂ surface. Ag nanoparticles with size below 5 nm are recommended for the inhibition of pathogenic microorganism growth [13]. Jialu Wu. et al. [14] examined the catalytic activity and stability of PtAg/TiO₂-rutile for the electrooxidation of methanol. Pt–Ag nanoparticles with an atomic ratio of 1:1 were prepared by the chemical co-reduction of precursors of Pt and Ag. Physical characterizations revealed that the Pt–Ag nanoparticles were evenly dispersed on the TiO₂. Pt,Ag/TiO₂-rutile showed significantly higher catalytic activity and stability than Pt,Ag/C, Pt/TiO₂, and Pt/C for methanol oxidation in both alkaline and acidic solutions, indicating that rutile TiO₂ is superior to carbon black as a support, and that Pt–Ag is superior to Pt in terms of catalytic activity. Also, M. Qamar [15] confirmed that the photocatalytic efficiency of TiO₂ nanoparticles in the degradation of acridine orange could be significantly improved by depositing the optimal amount of Pt and Ag metals. Panigrahi Puspamitra et al. [16] analyzed the electronic and optical properties of Pt–Ag-doped TiO₂, with different concentrations. The calculated reflectivity, energy loss function, and extinction coefficient demonstrated the enhanced optical properties of doped TiO₂ compared to its pristine form. These results underline the merits of tailoring the optical properties of Pt–Ag-doped TiO₂ for improved photocatalytic performance in the visible light region [16]. The above-mentioned works confirm that TiO₂ doped with Ag and Pt has higher photocatalytic activity than TiO₂ doped with only one of the metals. The size of the crystallites are also important for the optical and antibacterial properties of TiO₂. There are numerous works on TiO₂ coatings doped with Ag [4,17,18,19,20] or Pt [5,6,21]. We can also find works on TiO₂ coatings doped with Ag and Pt [13,14,15,16], but none describes a preparation via the magnetron sputtering process. This is due to the difficulty of embedding Ag and Pt nanoparticles in the TiO₂ matrix in the magnetron sputtering process.

Ag nanoparticles dispersed on the TiO₂ surface can significantly increase the photocatalytic and antimicrobial activity. Pt is one of the most active metals for photocatalytic enhancement under UV irradiation, producing the highest Schottky barrier among all metals and facilitating electron capture, thereby hindering the recombination between electrons and holes [13,15]. Due to its high photocatalytic and antibacterial activity, it was decided that the coating components should be TiO₂, Ag, and Pt.

Based on the above literature analysis, a TiO₂:Ag–Pt coating with very high fragmentation and excellent antibacterial effectiveness was successfully fabricated using magnetron sputtering. The crystal structure, hardness, Young’s modulus, surface geometric structure, wettability, and antimicrobial activity of TiO₂:Ag–Pt coatings produced via the magnetron sputtering process were determined and compared with those of TiO₂:Ag coatings produced by the same method [9,22]. The produced antibacterial TiO₂:Ag–Pt coating could be applied to protective glass glued to touch displays, such as those used in smartphones and smartwatches, as well as mirrors, glass panels in shower cabins (e.g., in hospital settings), and even surgical instruments [19].

2. Materials and Methods

In cooperation with A. Dziedzic from the University of Rzeszow and D. Augustowski and P. Kwasnicki from ML System S.A. Poland, the coating of TiO₂ doped with Ag and Pt was designed and prepared via a reactive magnetron sputtering process. The Minispectros thin-film deposition system (Kurt J. Lesker, East Sussex, UK), by ML System, was used to produce the TiO₂:Ag–Pt coating. The process chamber was equipped with two independent magnetrons operating in DC and RF modes. Original targets designed and manufactured by A. Dziedzic were used to produce the TiO₂:Ag–Pt coating. The TiO₂:Ag–Pt coating was deposited on a Si crystal and glass substrate. The sputtering source was two Ti targets manufactured by Kurt J. Lesker with a diameter of 3″, a thickness of 0.250″, and 99.995% purity. Titanium targets were sputtered simultaneously. To produce the coatings, six symmetrically arranged holes (diameter 2 mm) were drilled in the Ti target, and Pt or Ag pins were placed inside. The first titanium target with six Pt pins (purity of 99.9%) was sputtered using a DC power supply. The second titanium target with six Ag pins (2 mm diameter, purity of 99.9%) was sputtered using an RF power supply. The sputtering process was executed in a mixture of Ar/O₂ gases, with Ar having a purity of 99.9999%, and O₂ a purity of 99.999%. The O₂/Ar gases glow ratio was 0.3 (30%). Soda-lime glass plates (25 × 25 × 1 mm) and a Si crystal were used as the substrate. The substrates were cleaned in an ultrasonic cleaner in acetone and DI water and then dried with nitrogen. The sputtering process was carried out at a pressure of approximately 3.75 mTorr. The Ti target with Pt was sputtered using a power of 21.2 W/in², and that with Ag using a power of 26.5 W/in². Both the glass and Si substrates were heated to 300 °C during deposition coating in a vacuum chamber. The sputter process was carried out for 240 min. The TiO₂:Ag1.7, TiO₂:Ag2.0, and TiO₂:Ag2.5 coatings deposited on glass were prepared by sputtering one Ti target with six Ag pins. Applying a magnetron power of 7.1 W/in2, 10.7 W/in², and 14.3 W/in², Ag concentration in the coatings of 1.7% at., 2.0% at., and 2.5% at. were obtained, respectively.

The crystal structures of the coatings were analyzed by X-ray diffractometry (XRD) using a Bruker D8 Advance diffractometer (CuKα, λ = 0.154056 nm). To solve the resulting diffractograms, a base of ICCD PDF standards was used: anatase (00-021-1272), rutile (01-072-7374), Pt with a cubic crystal structure and face-centered lattice (01-087-0647), and Ag with a cubic crystal structure and face-centered lattice (00-004-0783). Angle and spot intensity matching were analyzed using the EVA software (ver. 5.2, PDF 2.1, Bruker Corporation, Billerica, MA, USA).

The structure was analyzed using a transmission electron microscope Tecnai Osiris, 200 kV (FEI Company, Eindhoven, The Netherlands). STEM HAADF (Z-contrast) imaging, HRTEM with FFT, and SAED diffraction images. Maps of the EDS chemical composition of elements were used to identify the crystallites from which the coating was composed. Lamella for TEM investigation was prepared using a focused ion beam microscope Quanta 3D 200i (FEI Company, Eindhoven, The Netherlands).

The surface and chemical composition of the coating were analyzed using a SEM microscope (FEI Quanta 3D 200i) with an EDS microanalyzer.

Surface morphology was probed by AFM (CSM) in contact mode. The roughness analysis consisted of identifying the root mean square (RMS) parameter.

Measurements of Young’s modulus and hardness were carried out using the nanoindentation method on a CSM stand, according to the ISO 14577 1:2015 09 standard [23]. The Young’s modulus of the tested coating was calculated based on the method of Oliver and Pharr [24]. The force acting on the indenter was 0.5 mN, the operating time was 15 s, and the rate of force increase was 1.0 mN/min. The maximum value of the loading force was selected so that the maximum impression depth was less than 10% of the coating thickness.

The sample transmission spectrum was determined at room temperature using a Cary 5000 spectrophotometer, Agilent Technologies, Santa Clara, CA, USA, Cary Win, software ver. 4.20 (468).

The water wetting angle of the coating was tested using a Nikon C-PS stereoscopic microscope with the camera from the MA-200 microscope (Nikon Corporation, Tokyo, Japan) and Nikon NIS Elements, software ver. 24.11.2013.

In the antibacterial tests, the coatings were activated with visible light, because in such conditions, the coatings would be used most of the time. For the antimicrobial assay, Gram-positive strain Staphylococcus aureus (ATCC 25923) was used. Nutrient broth and agar (BTL, Poland) or phosphate buffered saline (Sigma Aldrich, St. Louis, MI, USA, and Burlington, VT, USA) were used for bacteria culturing and serial dilutions, respectively. All assays were carried out in a laminar flow hood (BioTectum, Bielsko-Biala, Poland). The bacteria, having been incubated overnight in aerobic conditions at 37 °C with shaking, were diluted to give an initial concentration about 1 × 10⁴ cells/mL and poured onto the sterile coatings. During the incubations, samples of the coatings were irradiated with two 10 W LED lamps (3000 and 6000 K) which emitted visible light (400–700 nm) with an integral output of 0.5 mW/cm², resembling the light distribution of solar irradiation. After 2 and 4 h of incubation at a temperature 35 °C, the living bacteria were extracted from the sample surfaces using 5 mL SCDLP broth and ten-fold serial dilutions spread on the nutrient agar plates and incubated overnight at 37 °C. The results were determined by counting the colony forming unions (CFU). The percentage of growth reduction (%R) was calculation according to Formula (1):

%R = ((CFU control − CFU sample)/CFU control) × 100%

(1)

To analyze the bacterial morphologies, scanning electron microscopy (SEM) visualization was performed. The overnight growth of the S. aureus cell culture at 10⁸ CFU/mL was spread onto a surface and incubated for 4 h. After this, the samples were centrifuged and fixed with 2.5% glutaraldehyde for 2 h at room temperature. After fixation, the cell pellet was soaked in 0.1 M cacodylate sodium (pH 7) to remove excess fixative. The samples were then dehydrated by successive soakings in 25, 50, 75, and 95% ethanol for 10 min each. A thin layer of gold was then sputtered onto the samples, which were imaged using an SEM microscope.

Microbiological tests were carried out for the following coatings: TiO₂:Ag–Pt on Si, TiO₂:Ag–Pt on glass, and TiO₂:Ag on glass and antibacterial coatings for touch screens, purchased commercially.

2.1. XRD Analysis of the Structure

In order to determine the crystalline structure of the coating, XRD tests were carried out. Figure 1 shows the XRD patterns of the TiO₂:Ag–Pt coating produced on a Si crystal and soda-lime glass via a reactive magnetron sputtering process. Analysis of the obtained diffractograms of the TiO₂:Ag–Pt coating revealed the amorphous structure of the coating. No peaks from anatase, rutile, Pt, or Ag crystallites were identified in the diffractograms. Only small reflections coming from the substrate (Si crystal) are visible. However, our analysis using TEM methods, as shown in Section 2.2, confirmed that the coating, besides having an amorphous structure, consisted of crystallites of anatase, rutile, Pt, and Ag. Nanoparticles with sizes below 2–3 nm may not produce reflections in XRD patterns. The literature confirms that small crystal clusters are not detectable by XRD due to the lack of long-range order [25].

2.2. TEM Analysis of the Structure

In order to determine the thickness, structure, and phase composition of the TiO₂:Ag–Pt coating, TEM tests were carried out. Figure 2 shows an HAADF STEM image of a cross-section of the TiO₂:Ag–Pt coating. Thanks to the HAADF detector, we set the image contrast depending on atomic number Z of the elements, i.e., the higher the value of Z, the brighter the area. At the top of the image in Figure 2a, a Pt coating deposited in the process of preparing the thinned lamella using the FIB technique is visible. The structure of the coating comprised columnar crystals with a width of 30 to 50 nm, characteristic of production via the magnetron sputtering process. The columnar crystals crystallized in a direction perpendicular to the substrate and consisted of crystallites with different orientations, as revealed by HRTEM images (Figure 6). The coating deposited on Si had a thickness of 400 nm, and on glass 600 nm.

For the parameters adopted in this work, a coating with fine fragmentation was produced. The HAADF STEM image in Figure 3 shows Pt metallic nanoparticles (the brightest areas) and Ag nanoparticles (slightly darker areas) with sizes d < 3 nm. In the enlarged image in Figure 3, we observe that the nanoparticles formed aggregates with sizes up to 5 nm. A similar size of Ag–Pt crystallites was obtained by A. Zielińska-Jurek et al. [13] in Ag-Pt/TiO₂ photocatalysts produced by the sol-gel method (Pt approx. 3 nm and Ag approx. 6 nm). Also, M. Qamar [15] using the sol-gel method and photo-deposition, obtained Ag and Pt nanoparticles on a TiO₂ surface with sizes of 2–4 nm and 3–20 nm, respectively. Adochite R.C. et al. [21] obtained Ag particles with a size of approx. 5 nm in a TiO₂ matrix immediately after deposition. In the works mentioned above, Ag and Pt nanoparticles of significantly different sizes were obtained. In our study, the produced TiO₂:Ag–Pt coating consisted of crystallites in a rather narrow size range, i.e., up to 3 nm. The large number of small-sized Ag and Pt particles generated a large surface area of nanoparticle boundaries, which translated into increased antibacterial properties [13]. The presence of metallic Ag–Pt nanoparticles in the ceramic matrix caused the formation of plasmon resonance, which increased the photocatalytic efficiency; the smaller the nanoparticles, the higher the photocatalytic efficiency [26,27,28].

The diffraction pattern of the TiO₂:Ag–Pt coating (Figure 4) shows reflections from rutile R(110), R(200), Pt(200), and Pt(111). A wide ring from the amorphous structure is also visible.

Figure 5 shows the element distribution maps of a cross-section of the TiO₂:Ag–Pt coating. The maps reveal a homogeneous distribution of Ti and O forming the ceramic matrix and agglomerates of Ag and Pt metal nanoparticles.

In order to identify the crystallites that formed the TiO₂:Ag–Pt coating and determine their sizes, HRTEM imaging was performed (Figure 6). HRTEM images revealed the crystalline structure of the coating. Fourier transform was performed on the areas marked with a red square. Then, the distances between the atomic planes were calculated (Figure 6). Based on the determined distances, crystallites were identified. Crystallites of anatase A(101), A(103), rutile R(101), R(200), R(111), (R(210), Ag(111), Ag(200) and Pt(111), and Pt(200) were identified in the TiO₂:Ag–Pt coating deposited on Si. However, in the TiO₂:Ag–Pt coating deposited on glass, crystallites of anatase A(101), A(103), A(112), rutile R(101), R(200), Ag(111), and Pt(111) were observed. The crystallite size did not exceed 3 nm. HRTEM images mainly show agglomerates of the same and different crystallites.

XRD diffractograms, HAADF STEM, HRTEM images, SAED patterns, and chemical composition maps confirmed that a nanocomposite coating had been obtained as a result of reactive sputtering. It was a nanocomposite with a ceramic crystalline and partially amorphous TiO₂ matrix and metallic crystalline Pt and Ag nanoparticles. The crystallite size did not exceed 3 nm. An analysis of the literature [13,15,21] shows that this is a very large fragmentation of the structure.

2.3. SEM with EDS Analysis

In order to determine the surface topography and chemical composition of the TiO₂:Ag–Pt coating, tests were performed using a SEM microscope equipped with a composition microanalyzer. The results revealed the presence of Pt and Ag particles on the surface of the TiO₂:Ag–Pt coating, with sizes reaching up to 150 nm (Figure 7). Ag particles of similar size on the surface of coatings were observed by S. Calderon Velasco et al. [17]. Metal particles on the coating surface were formed not only as a result of agglomeration but also through the phenomenon of coalescence, as well as through the dissolution of smaller particles and the formation of larger particles [17]. It should be noted that during the production of the TiO₂:Ag–Pt coating in the vacuum chamber, the substrates were heated to a temperature of 300 °C. Without heating the coating, no anatase was obtained, only rutile. Anatase is more desirable due to its greater degree of surface hydroxylation [19]. An analysis of the chemical composition allowed us to conclude that the TiO₂:Ag–Pt coating on the glass contained 5.0 ± 0.2% at. Ag and 4.4 ± 0.1% at. Pt. The coating prepared on Si had a higher concentration of metallic elements, i.e., 9.1 ± 0.2% at. Ag and 5.7 ± 0.2% at. Pt.

2.4. Analysis of the Geometric Structure of the Coating Surface

To assess the quality of the TiO₂:Ag–Pt coating, the geometric structure and RMS roughness were examined using atomic force microscopy (AFM). Figure 8 and Figure 9 show 2D and 3D maps and profiles of the surface of the TiO₂:Ag–Pt coating on Si crystal and glass. The surface topography of the TiO₂:Ag–Pt coating on Si and glass yielded similar AFM maps of the surface, as well as TEM images, confirming that the TiO₂:Ag–Pt coating consisted of columnar crystals with a diameter of 30–50 nm. AFM images showed grains of this size. Columnar crystals, as they grow, increase slightly in diameter [17]; hence, we can observe a slightly larger crystal diameter in the AFM maps than in the TEM images. The RMS surface roughness for the TiO₂:Ag–Pt coating deposited on Si was RMS = 2.6 ± 0.3 nm and on glass RMS = 1.8 ± 0.2 nm.

2.5. Mechanical Properties Tests

In order to determine the durability, the hardness and Young’s modulus of the TiO₂:Ag–Pt coating were tested. An analysis of the measurement results showed that the hardness of the coating deposited on Si was HV = 6.5 ± 0.5 GPa and on glass HV = 6.8 ± 0.5 GPa. A slight difference in coating hardness may have resulted from differences in the Ag and Pt concentrations in the coatings, depending on the substrate. The Young’s modulus of the coating on Si was E = 124 ± 6 GPa and on glass E = 116 ± 8 GPa. The higher value of the Young’s modulus for the coating deposited on Si may have resulted from the larger Young’s modulus for Si E = 120 ± 6 GPa. The measured Young’s modulus for the glass was E = 70 ± 5 GPa. Figure 10a,b show the indenter load curve as a function of the indentation depth for a Si substrate and a TiO₂:Ag–Pt coating deposited on Si and for a glass substrate and a TiO₂:Ag–Pt coating deposited on glass. A noticeable difference in the increase of the indenter load was observed between the glass and Si substrates, which could be attributed to the distinct properties of these materials.

2.6. Wettability Tests

The following coatings were tested: TiO₂Ag–Pt on glass, TiO_2:Ag–Pt on Si, TiO₂:Ag1.7 on glass, TiO₂:Ag2.0 on glass, and TiO₂:Ag2.5 on glass. Water contact angle tests showed that the TiO₂:Ag–Pt coating was characterized by hydrophilic properties. After being sprinkled with water, the TiO₂:Ag–Pt coating had a contact angle of 77.2 ± 7.3° (Figure 11). After 3, 6, 9, 12, and 15 min, the angle decreased to 64.2 ± 6.7°, 49.6 ± 2.9°, 40.1 ± 3.1°, 25.6 ± 3.2°, and 13.5 ± 3.4°, respectively. For comparison, the authors also analyzed the wettability of TiO₂:Ag coatings. The tests showed that TiO₂:Ag coatings had a contact angle smaller than the TiO₂:Ag–Pt coatings. It was observed that with an increase in the concentration Ag in the TiO₂:Ag coating from 1.7 to 2.5% at., the contact angle increased. The TiO₂:Ag coating with a Ag concentration of 1.7, 2.0 and 2.5% at. had a contact angle of 38.0 ± 6.0°, 42.4 ± 5.8°, and 55.9 ± 3.0°, respectively. After 15 min, the contact angle decreased for all TiO₂:Ag coatings. By increasing the Ag concentration from 2.1 to 2.9% at., the contact angle also increased, as confirmed by A. Jagodzińska [29]. Mosquera [18] also confirmed that with an increase in Ag content from 0.08 to 0.28% at., the contact angle increased with TiO₂:Ag coatings. This means that the coatings were less hydrophilic [18]. However, Fanming Meng et al. [4] claimed that the hydrophilic properties of a foil increased with an increase in Ag concentration up to 5% vol. Ag and then decreased [4]; this means that the water wetting angle of the films decreased as the Ag concentration increased to 5% vol. Ag and then increased. The TiO₂:Ag–Pt coating contained more Ag 9.1 ± 0.2% at. and also contained Pt 5.7 ± 0.2% at. Sometimes, a higher contact angle can be explained by a higher roughness of the coating, but in this case, the TiO₂:Ag coatings had an RMS roughness of 3.6–5.3 nm, while the TiO₂:Ag–Pt coating had an RMS roughness of 1.8–2.6 nm. However, the presence of Pt in the coating may have changed the wetting mechanism. Undoped titanium dioxide is characterized by photoinduced superhydrophilicity of the surface. As such, it is used in the production of self-cleaning and anti-fog glass and self-cleaning coatings [30].

Doping a semiconductor with metals alters its surface properties, modifies the ratio of carrier concentrations at the surface, and affects the composition of the hydroxyl-hydrated layer. These changes consequently influence the initial hydrophilic state of the surface and its photoinduced transformation [31].

2.7. Optical Properties of Coatings

The transmission spectrum of the TiO₂:Ag–Pt layer on a glass substrate, measured in the range of 200–800 nm, is shown in Figure 12.

In the short wavelength range, i.e., below 350 nm, the layer was impermeable. Then, an increase in layer transmission with wavelength was observed, with interference minima and maxima testifying to the high homogeneity of the layer thickness [32]. In the range of 500–800 nm, the average transmission value was 66.4%, so slightly smaller than that of the layers of TiO₂ on a glass substrate (with similar thickness), as described in the literature [33].

Based on the transmission of the layer, its absorption coefficient was determined in accordance with Formula (2):

$$\mathsf{\alpha }={\displaystyle \frac{1}{\mathrm{d}}}\mathrm{l}\mathrm{n}\left({\displaystyle \frac{1}{\mathrm{T}}}\right)$$

(2)

where d is layer thickness (in cm) and T is transmission.

In the area of transparency, the layer of the absorption coefficient value did not exceed several thousand cm⁻¹, while around the fundamental absorption edge, a value of 3 × 10⁵ cm⁻¹ was reached, which shows compliance with literature data for TiO₂ films [34].

In order to determine the value of the optical band gap energy, the Tauc model was used, using the relationship [35,36] (3):

$$\left({\mathsf{\alpha }\mathrm{h}\mathrm{v}}^{\mathrm{m}}\right)=\mathrm{A}\left(\mathrm{h}\mathrm{v}-{\mathrm{E}}_{\mathrm{g}}\right)$$

(3)

where A is constant, E_g is the optical band gap energy, and m = 2 for direct and 0.5 for indirect transitions.

In the literature, there is no consensus about the type of transitions for TiO₂, with researchers reporting both the occurrence of direct [35] and indirect [36] transition.

In our case, the model of direct transition (m = 2) turned out to be better. The spectrum of absorption in the Tauc coordinates (for the case direct transition) for the tested layer is shown in Figure 13.

The value of the obtained optical band gap energy corresponding to the direct transitions was 3.36 eV. According to the literature, for thin layers of TiO₂, E_g is in a wide range of 3.0–3.8 eV [37,38,39,40,41,42], depending on TiO₂ phase. For the anatase (~3.2 eV) phase, it was greater than rutile (~3.0 eV), and for the amorphous phase greater than for crystalline.

The Urbach tail, which occurs for disordered, poorly crystalline and amorphous materials, is an exponent part of absorption appearing in the area of the fundamental absorption edge. It is described by the dependence [43] (4):

$$\left\{\begin{array}{c}\mathsf{\alpha }\left(\mathrm{h}\mathrm{v}\right)={\mathsf{\alpha }}_0\exp \left({\displaystyle \frac{\mathrm{h}\mathrm{v}}{{\mathrm{E}}_{\mathrm{u}}}}\right)\\ {\mathrm{E}}_{\mathrm{U}}=\mathsf{\alpha }\left(\mathrm{h}\mathrm{v}\right){\left[{\displaystyle \frac{\mathrm{d}\left[\mathsf{\alpha }\left(\mathrm{h}\mathrm{v}\right)\right]}{\mathrm{d}\left[\mathrm{h}\mathrm{v}\right]}}\right]}^{-1}\end{array}\right.$$

(4)

where E_U is the Urbach energy, hv is the photon energy, α₀ is a constant, and α is absorption.

Figure 14 shows the dependence of ln(α) on the energy of falling photons.

The determined E_U value was 508 meV, i.e., within the value range for TiO₂ reported in the literature (254–806 meV [44,45]). It should be noted that the increase in the concentration of oxygen defects in the layer led to the greater E_U value [45]).

2.8. Antibacterial Testing of Coatings

The prepared TiO₂:Ag–Pt coatings on Si and glass were subjected to microbiological tests. For comparison, a commercial antibacterial coating on protective glass for touch screens and a TiO₂:Ag coating on glass were analyzed. After 2 h of incubation of bacteria on the surface of the tested plates, the tests showed a bacterial reduction of 20.0%, 30.0%, 86.7%, and 97.7% for the commercial coating, TiO₂:Ag, TiO₂:Ag–Pt on Si, and TiO₂:Ag–Pt on glass, respectively (Figure 15a). After 4 h of incubation, bacterial reduction was 59.1%, 55.4%, 94.5%, and 90.3% for the commercial coating, TiO₂:Ag, TiO₂:Ag–Pt on Si, and TiO₂:Ag–Pt on glass, respectively. The TiO₂:Ag–Pt coating on glass was characterized by the highest reduction of S. aureus bacteria, i.e., above 90%, regardless of the incubation time. The TiO₂:Ag coating on glass was characterized by a similar maximum bacteria reduction to that of TiO₂:Ag,N coatings [9], which confirms the correctness of the research methodology used. The commercial coating was characterized by a level of bacteria reduction similar to the TiO₂:Ag coating. Figure 15b shows exemplary dishes with colonies of S. aureus bacteria from glass control plates (left) and from the TiO₂:Ag–Pt coating on Si (right). Several bacterial colonies from the TiO₂:Ag–Pt coating and several hundred colonies from the control plate are visible. The more than 90% reduction of S. aureus bacteria on the TiO₂:Ag–Pt coating on glass was caused by the unique phase composition, metallic Ag–Pt nanoparticles in the ceramic matrix of anatase and rutile crystallites, and a small volume fraction of amorphous structure. The crystallites size did not exceed 3 nm. This combination of phase composition ensured over 90% reduction of bacteria thanks to the activation of the coating with an electromagnetic wave in the visible range and additional strengthening thanks to the phenomenon of plasmon resonance of Pt and Ag nanoparticles. The increase in photocatalytic efficiency via plasmon resonance of Ag–Pt nanoparticles was confirmed by M. Endo-Kimura et al. [46]. Those authors also proved that plasmonic photocatalysts are effective agents against various microorganisms. Moreover, regardless of light, the coating released Ag⁺ ions [17,19], also killing bacteria.

The bacterial morphologies of representative samples taken after incubation with every plate were examined by SEM (Figure 16). It can be seen that the bacteria in the control group or those exposed to commercial glass had smooth surfaces and maintained the integrity of membrane; the cells were spherically shaped. In contrast, after contact with the modified plate (TiO₂:Ag–Pt on glass and TiO₂:Ag–Pt on Si), cells exhibited profound morphological deformations. The cell damage and shrunken appearance indicated a loss of structural integrity. Leakage of the cytoplasmic components can be also observed. Peng-Fei Cai et al. reported on a photocatalytic TiO₂:Ag thin film and its sterilization potential against S. aureus and E. coli. The antibacterial efficiency was around 98% according to their examination of the morphological changes of the treated bacteria cells [47].

3. Conclusions

A TiO₂ coating doped with Ag and Pt was successfully produced by reactive magnetron sputtering. The TiO₂:Ag–Pt coating had a columnar structure (column width—30–50 nm) and consisted of a ceramic crystalline and amorphous matrix of rutile and anatase and metallic crystalline Pt and Ag nanoparticles. The crystallite size did not exceed 3 nm. Nanoparticles with diameters up to 150 nm were identified on the coating surface. The TiO₂:Ag–Pt coating deposited on glass had a concentration of 5.0 ± 0.2% at. Ag and 4.4 ± 0.1% at. Pt. The hydrophilic coating had a roughness RMS = 1.8 ± 0.2 nm, hardness HV = 6.8 ± 0.5 GPa, and Young’s modulus E = 116 ± 8 GPa. The contact angle for the TiO₂:Ag–Pt coating was 77.2 ± 7.3°, decreasing over time to 13.5 ± 3.4° (after 15 min). The TiO₂:Ag–Pt coating had a contact angle larger than those of the TiO₂:Ag coatings. Optical studies of the TiO₂:Ag–Pt layer on a glass substrate showed that the optical properties (e.g., absorption coefficient, E_g, E_U) were close to those given in literature for TiO₂ layers. The prepared TiO₂:Ag–Pt coating showed a high, i.e., over 90%, reduction of S. aureus bacteria, while TiO₂:Ag and commercial coatings achieved a 60% reduction of bacteria. The TiO₂:Ag–Pt coating deposited on glass and on Si had comparable wetting angles and were characterized by a similar reduction of bacteria, i.e., within the limits of measurement uncertainty. The observed properties make the TiO₂:Ag–Pt coating suitable for use on protective glass glued to touch displays.