Antibacterial Activity of PVA Hydrogels Embedding Oxide Nanostructures Sensitized by Noble Metals and Ruthenium Dye

Nanostructured oxides (SiO2, TiO2) were synthesized using the sol–gel method and modified with noble metal nanoparticles (Pt, Au) and ruthenium dye to enhance light harvesting and promote the photogeneration of reactive oxygen species, namely singlet oxygen (1O2) and hydroxyl radical (•OH). The resulting nanostructures were embedded in a transparent polyvinyl alcohol (PVA) hydrogel. Morphological and structural characterization of the bare and modified oxides was performed using scanning electron microscopy (SEM), transmission electron microscopy (TEM), atomic force microscopy (AFM), UV–Vis spectroscopy, and X-ray photoelectron spectroscopy (XPS). Additionally, electrokinetic potential measurements were conducted. Crystallinity data and elemental analysis of the investigated systems were obtained through X-ray diffraction and X-ray fluorescence analyses, while the chemical state of the elements was determined using XPS. The engineered materials, both as simple powders and embedded in the hydrogel, were evaluated for their ability to generate reactive oxygen species (ROS) under visible and simulated solar light irradiation to establish a correlation with their antibacterial activity against Staphylococcus aureus. The generation of singlet oxygen (1O2) by the samples under visible light exposure can be of significant importance for their potential use in biomedical applications.


Introduction
In the last few decades, the global economic boom has triggered a tremendous search for better healthcare solutions and the preservation of quality of life, especially in densely populated communities where intensive natural resource usage has had an impact. One of the key challenges that emerged during this time was the increasing resistance of bacteria to antibiotics and the aggressive nature of cancer. Overcoming these challenges requires the validation of new antibacterial and theranostic agents.
Extensive research has been conducted to explore the interaction between inorganic materials, such as oxides and metals with endogenous compounds and living systems. The aim is to develop biocompatible nanostructures that can serve as safe implants or prosthetics further modified with ruthenium dye and noble metal nanoparticles (Au and Pt), with glutathione serving as a capping agent. The functional characterization of the synthesized samples was performed by monitoring the photogeneration of reactive oxygen species (hydroxyl radical and singlet oxygen) and conducting antibacterial assays against S. aureus in both dark and light-exposed conditions. Figure 1a,b show SiO 2 tubes reaching hundreds of micrometers in length and variable diameters from nano to micrometric range together with spherical particles located on their external surface, in line with our previously reported data [33][34][35]. Together with tubular morphology, the other structural and functional particularities (optical activity, singlet oxygen generation) of this atypical SiO 2 are expected to be displayed. Figure 1c,d present the TiO 2 nanoparticles about 20-40 nm, sticking together and developing big aggregates.

AFM Characterization
AFM investigations ( Figure 3) were conducted to explore the morphology of the samples not characterized by other methods, namely: the hydrogel film and the free Au, Pt nanoparticles NPs (unsupported on SiO 2 and TiO 2 matrices). Figure 3. Topographic 2D AFM image, scanned over an area of (2 µm × 2 µm) of the prepared hydrogel, (a) together with a random height vs. distance plot (b); of the AuNPs (c) accompanied by a line-scan characteristic for the AuNPs (d); of the PtNPs (e) alongside with the corresponding profile (line-scan) of the PtNPs (f). Figure 3a presents the morphology of the prepared hydrogel, as seen in topographic images recorded by AFM at the scale of (2 µm × 2 µm). It can be observed that the surface of the hydrogel consists of a random agglomeration of material in the form of hills alternating with valleys (inter-hills free spaces), resulting in a height difference of a few tens of nanometers in the z-direction. For example, the line scan presented in Figure 3b suggests a height difference of approximately 25 nm (ranging from −10 to +15 nm). The entire scanned area shown in Figure 3a exhibits the following corrugation parameters: an RMS roughness (R q ) of 11.6 nm and a peak-to-valley (R pv ) of 147.4 nm. Figure 3c,e show the morphology of the Au and Pt nanoparticles. Both samples were prepared by their dispersion in ultra-pure water, followed by drop deposition on clean microscopic glass substrates. The AuNPs, exhibited in Figure 3c, have particles with diameters ranging from a few tens of nanometers, such as 30 nm, up to approximately 200 nm. However, most of the AuNPs shown in Figure 3c have a diameter of around 100 nm (consistent with the corresponding line scan plotted in Figure 3d. The PtNPs (Figure 3e) tend to cluster in bunches with cauliflower-like structures, with diameters in the range of 200-300 nm. The constituent particles within these bunches have a mean diameter of less than 50 nm (Figure 3e), also noticeable in the line profile from Figure 3f, indicating that each large particle (each peak) is composed of smaller neighboring particles, resulting in a "modulated" profile. The AFM characterization of Au and Pt-modified TiO 2 and SiO 2 powders is presented in Figures S1-S3 from Supplementary Materials.

XRD and XRF Characterization
XRD and XRF analyses were performed to confirm the presence of the dopants in the TiO 2 and SiO 2 matrices, revealing the existence of Au and Pt crystalline phases in the samples. Figure 4a,b display the diffractograms of Au-doped and Pt-doped samples, respectively. Two crystalline phases were observed in the Au-doped TiO 2 (AuTiO 2 ) sample: metallic Au and anatase TiO 2 . The Au and anatase phases were identified based on the ICDD file no. 4-0784 and no. 21-1272, respectively. The RIR method was employed to calculate the quantity of Au relative to anatase TiO 2 and was estimated to be 1.92%. The Au-doped SiO 2 sample (AuSiO 2 ) contains a metallic Au phase and an amorphous SiO 2 matrix. The PtSiO 2 sample, on the other hand, turned out to be similar to the AuSiO 2 sample, with an amorphous SiO 2 matrix and a metallic Pt phase detected by XRD using ICDD file no. 4-0802. The amount of platinum identified by XRF analysis was 0.46% by mass. The Pt-TiO 2 sample was noted to have a complex composition, with the identification of anatase TiO 2 , Pt, and two sodium titanates (Na 2 Ti 6 O 13 and Na 4 Ti 5 O 12 ) based on ICDD files nos. 14-0277 and 37-0273, respectively. Additionally, a small amount of NaCl was detected. The XRF analysis of the PtTiO 2 sample indicated a platinum quantity of 0.74% by mass. According to the XRD analysis, the crystallite sizes of metals are: 10 nm AuTiO 2 , 27 nm AuSiO 2 , 18 nm PtTiO 2 , and 14 nm PtSiO 2 .      The registered XPS lines are assigned according to Table 1.

UV-Vis Characterization
In order to reveal the light absorptive properties of the investigated materials, the UV-Vis characterization was performed on SiO 2 and TiO 2 samples before and after their modification as well as for PVA-based gels containing the mentioned powders. Accordingly, the UV-Vis spectra were recorded for the following samples (Table 2). The spectra were recorded between 200 and 1000 nm (Figure 6a,d) and represent the gel-containing samples in the 200-400 nm (Figure 6b,e) and 400-1000 nm (Figure 6c,f) for a better evaluation of their light absorptive properties in the UV and Vis domains. Tubular SiO 2 (SiO 2 sample) shows a broad absorption band spanning from the UV to the visible domain (300-600 nm, Figure 6a), which is amplified by modifiers and redshifted. The presence of gold (AuSiO 2 sample) induces the appearance of the surface plasmon resonance phenomenon (peak centered at 550 nm), while the addition of ruthenizer (RSiO 2 sample) results in a maximum absorption peak around 520 nm. Furthermore, platinum nanoparticles (PtSiO 2 sample) significantly enhance the light absorption capacity of the sample. The TiO 2 sample exhibits an absorption edge at around 400 nm. These considerations also apply to the TiO 2 -based samples, where the presence of gold (AuTiO 2 sample) and ruthenizer (RTiO 2 sample) is marked by well-defined peaks in the visible domain. In the case of gold, there is a slight red shift relative to SiO 2 .
Except for the silica modified with ruthenizer and embedded in gel (GRSiO 2 sample), the light absorption in the UV domain of the gel-containing powders occurs between 200 and 250 nm and is lower (Figure 6b,e) than for the bare gel (Gel sample) but higher than for the oxide powders in their standalone form. Although the PVA-DMSO (dimethyl sulfoxide) gel proved significant transparency in the range of 400-700 nm [11], the powders embedded in gels show no light absorption in the visible domain (Figure 6c,f).

i. Generation of hydroxyl radicals (•OH) under light irradiation
The presence of hydroxyl radicals (•OH) was checked in a coumarin aqueous solution containing the samples of interest (powders or gels) and exposed to simulated solar light. The same tests have been performed for visible light irradiation, but all the investigated samples proved to be without activity. In the case of solar light utilization, some active samples generate (•OH) that leads after reaction with coumarin to the umbelliferone formation, a fluorescent product signaled by a PL emission at 451 nm for λexc = 330 nm [34].
No significant activity for the SiO 2 -based samples was registered, with only small amounts of hydroxyl radicals being produced after 30 min of irradiation by the AuSiO 2 powder sample ( Figure 7a). Also, the GSiO 2 sample (SiO 2 embedded in PVA gel) proved a slight tendency in this sense, probably due to the gel properties (Supplementary Information- Figure S4). TiO 2based samples show a higher capacity to generate hydroxyl radicals under solar irradiation, according to the sequence TiO 2 > PtTiO 2 >RTiO 2 > AuTiO 2 (Figure 7b-e) but this disappears after mixing with gel (Supplementary Information- Figure S4). The PVA gel produces only traces of •OH (Figure 7f).
Hydroxyl radical is known as a powerful oxidant agent, acting on various organic biomolecules [12,36]. Accordingly, the identified samples as •OH providers under solar light exposure could act as antibacterial agents too.
ii. Generation of Singlet Oxygen ( 1 O 2 ) under visible irradiation (λ > 420 nm) monitored by using Singlet Oxygen Sensor (SOSG) λexc = 480 nm The generation of oxygen singlet particularly for applications in the biomedical field, especially such as photodynamic therapy, has received significant interest. The target samples, when exposed to a methanolic solution of SOSG, produce photogenerated oxygen singlet ( 1 O 2 ), leading to the formation of endoperoxide. This process is indicated by a peak in photoluminescence (PL) emission around 530 nm. The time course of 1 O 2 generation by SiO 2 powder is illustrated in Figure 8a, clearly showing an increasing amount over time. While the addition of ruthenizer does not contribute to singlet oxygen generation (Figure 8b), the incorporation of gold greatly enhances this process (Figure 8c).
The oxygen singlet generation by the highly defected SiO 2 was previously reported by our group [35] but the discovery of its improvement through the addition of gold is a novel finding. Unlike this, modification with gold of the TiO 2 does not significantly change its activity (Figure 8f) but a small increase in 1 O 2 generation is obtained for the ruthenizer-modified TiO 2 (RTiO 2 sample) (Figure 8e). Our previous work also reported the TiO 2 ability to generate oxygen singlet under visible irradiation without improvements brought by Au and Ag nanoparticles [37].
Since 1 O 2 induces bio-membrane degradation [12], a correlation of these data with the antibacterial activity is of great importance for practical applications. The PVA (gel sample) shows a significant activity for 1 O 2 generation (Figure 8g).

Electrokinetic Potential Measurements
Electrokinetic potential measurements are meant to investigate the charged surface of the engineered nanoparticles in water and the presumable interaction with bacterial cell surface that usually bears a negative net charge at pH 7. Table 3 shows the negative zeta potential for both oxide powders (SiO 2 and TiO 2 ) that shift positively by mixing with gel whose zeta potential is quite close to 0. The experiments were made in triplicates. The modification of SiO 2 with metals results in a decrease in zeta potential values, with platinum-modified SiO 2 (PtSiO 2 ) reaching as low as −60 mV. In contrast, the addition of metals to the TiO 2 samples increases the potential values. Based on these findings, it can be inferred that the surface charge of the gel and gel containing TiO 2 powders is closest to establishing contact with bacterial cells.

Antibacterial Activity Assays of the Investigated Samples against S. aureus 2.8.1. Antibacterial Activity Assays in Dark
The antibacterial activity assay of the investigated samples against S. aureus is presented in Figure 9. Figure 9a allows comparing the antimicrobial activity of the investigated samples in the dark, leading to a hierarchization of the target materials. Both unmodified SiO 2 and TiO 2 samples show insignificant antibacterial activity (Figure 9a) which is consistent with the reported data [24]. The lowest cellular viability relative to the control (S. aureus) was registered for the PtTiO 2 sample, the gel embedding of PtTiO 2 powder has a lower antibacterial effect. Therefore, the notable reduction in microbial growth induced by the Pt-TiO 2 sample (Figure 9a) can be assigned to the presence of platinum nanoparticles in the TiO 2 matrix and to their interaction. The antibacterial activity of platinum nanoparticles against Gram-negative and Gram-positive bacteria (such as E. coli and S. aureus) is well-documented [15,16,18,38]. Ahmed et al. [17] reported the use of PtNPs 2-5 nm in size for reducing bacterial cell viability through reactive oxygen species (ROS) production and membrane integrity loss. But, according to Aygun et al. [18], ROS formation can also be induced in mammalian cells by the platinum nanoparticles, and their use for therapeutic purposes requires more data.
Supplementary data are provided in Figure 9b concerning the antibacterial effect of PtTiO 2 -based materials in the dark. A sharp decrease in logarithm in colony-forming units is depicted for PtTiO 2 powder when increased the inorganic material amount (from 0.001-0.004 g). Also, the gel sample proves intrinsic antibacterial activity, therefore enhancing the antibacterial effect of the embedded PtTiO 2 powder. Table 3 indicates a negative electrokinetic potential for the PtTiO 2 material, suggesting that bacterial cell adhesion to the PtTiO 2 sample could be hindered. Thus, it is reasonable to assume that other antibacterial mechanisms, such as reactive oxygen species (ROS) generation and ion release, are at play. Unlike this, the antibacterial effect of PtSiO 2 material is almost insignificant, the morphological and structural properties of the deposited metal nanoparticles proving to be dependent on the support characteristics. Although it has lower antibacterial reactivity, AuTiO 2 seems to be closer to PtTiO 2. The unsupported Au nanoparticles also emphasize antibacterial activity in the dark, the accumulation of AuNPs on the cell wall, their diffusion, and the resulting cell lysis is more pronounced in Gram-negative (E. Coli) than in Gram-positive bacteria (S. aureus) [15,39].

Antibacterial Activity Assays under Light Exposure
According to Figure 10a, bacterial cell viability is primarily reduced in the presence of a TiO 2 sample exposed to solar light irradiation. This aligns with the findings in Figure 7b which confirms the ability of TiO 2 to photogenerate hydroxyl radicals. The formation of singlet oxygen is also conceivable, considering that the cell culture medium contains both D 2 O and H 2 O. The TiO 2 -based powders exhibit superior activity compared to the SiO 2 ones. While adding gel to bare TiO 2 does not provide any benefits, it leads to a slight decrease in cell viability for its derivative samples (GAuTiO 2 , GRTiO 2 ). The gels containing SiO 2 -based materials show better activity than free powders under solar light exposure.   Figure 8a,c confirms that SiO 2 but especially the AuSiO 2 sample produce oxygen singlet when exposed to visible light. The enhanced singlet oxygen production was achieved using D 2 O instead of H 2 O, which generally acts as a quencher for singlet oxygen. By incorporating gold nanoparticles as modifiers in the SiO 2 matrix, an improvement in light absorption and activation of surface defects was achieved, leading to increased singlet oxygen production as the primary antibacterial mechanism. The ruthenizer used as a sensitizer in the case of TiO 2 samples proves to be efficient too.
Comparing Figure 10a,b, a better antibacterial activity of the bare SiO 2 appears under visible compared to solar light irradiation. This may initially seem contradictory since, by exposure to solar irradiation, SiO 2 can utilize the light energy in the UV range (corresponding to the solar spectrum), not only from the visible spectrum. However, this result can be explained by the higher production of singlet oxygen, favored during the tests containing only D 2 O (without water). Unlike this, under solar light exposure, hydroxyl radicals are predominantly generated, which explains the higher activity of the TiO 2 sample (Figure 10a). The antibacterial assays conducted under visible light exposure and depicted in Figure 10b, indicate the addition of gold to silica as the best modification for oxide matrices. Despite the enhanced light absorption, the addition of ruthenizer does not significantly reduce microbial growth observed for SiO 2 . In the case of TiO 2 , the decrease in cell viability with the increase in the modified sample (0.001-0.004 g RTiO 2 ) can be due to the presence of a ruthenizer.
Therefore, several additional considerations can be made regarding the AuSiO 2 sample when exposed to light. UV-Vis measurements reveal that the bare oxides (SiO 2 and TiO 2 ) exhibit light absorption properties that cover the UV-Vis and UV domains. The modification of these matrices with metals increases light absorption and induces surface plasmon resonance, consistent with the reported literature [40,41]. The presence of gold is observed in both SiO 2 and TiO 2 but its effect on enhancing oxygen singlet generation is only evident in the case of Au-SiO 2 sample. This observation aligns well with the antibacterial activity exhibited by the AuSiO 2 sample against S. aureus under visible light irradiation. In this case, the antibacterial assay was conducted in the presence of D 2 O, without H 2 O, to specifically highlight the formation of singlet oxygen. Consequently, the better antibacterial effect observed for the AuSiO 2 sample can be attributed to the presence of singlet oxygen which is promoted by the presence of gold in the SiO 2 matrix. The bioactivity of gold nanoparticles is well-established and extensively studied [42,43]. This result is significant as it demonstrates the improvement in oxygen singlet photogeneration compared to the bare sample and it is achieved under visible light irradiation, unlike similar studies that employ UV irradiation for both SiO 2 and TiO 2 [12].
Although many silica-based materials have been tested for their antimicrobial properties [44,45], various modifiers have been used since the bare SiO 2 did not indicate intrinsic antimicrobial properties, neither in dark [46][47][48] nor under light exposure [49]. Based on these considerations, it can be assumed an advance in the biomaterials field triggered by this result: the endowing of an inactive material, namely SiO 2 , with antibacterial properties by exposing it to visible light irradiation. More than that, the antibacterial capacity can be increased by modifying the SiO 2 matrix with gold nanoparticles due to the singlet oxygen generation. This aspect is all the more important as it is known the ability of metal nanoparticles to induce oxidative stress on the bacterial cell, the singlet oxygen is responsible to a greater extent for cell damage than the super anion radical and hydrogen peroxide (endogenous antioxidants can easier reduce their effect) [19].

Conclusions
The present work highlights the development of an efficient antibacterial agent against S. aureus in the dark, specifically PtTiO 2 powder. This goal is achieved by modifying previously synthesized sol-gel TiO 2 nanoparticles with glutathione-capped platinum nanoparticles. Additionally, the obtained PVA hydrogel exhibits intrinsic antibacterial activity and acts as an effective dispersive medium for the PtTiO 2 powder, enhancing its antibacterial behavior.
The results of antibacterial assays conducted under visible light irradiation reveal the potential of highly defected SiO 2 with tubular morphology as an optically active material capable of reducing the cell viability of S. aureus through the photogeneration of singlet oxygen. Moreover, the antibacterial properties of SiO 2 can be further improved by modifying its matrix with glutathione-capped gold nanoparticles.
These findings represent significant progress towards SiO 2 and AuSiO 2 applications in the biomedical field (e.g., photodynamic therapy) owing to their ability to generate singlet oxygen under visible light irradiation.

Synthesis of SiO 2 and TiO 2 Matrices
SiO 2 with tubular morphology was synthesized by using a modified sol-gel method, according to previously reported works [33][34][35]. DL tartaric acid (TA, Riedel de Häen, Buchs SG, Switzerland) was dissolved in ethanol absolute (99.5% Merck) and ultrapure water. For the template formation, namely ammonium tartrate, gaseous ammonia was bubbled in the above-mentioned mixture. TEOS (tetraorthosilicate, 99%, AlfaAesar) was slowly added according to the molar ratio: 1 TEOS/0.035 TA/21.5 C 2 H 5 OH/18 H 2 O. The resulting gel was filtered, dried, and thermally treated in air at 500 • C for 3 h.
TiO 2 nanoparticles were obtained by sol-gel method using titanium isopropoxide (97%, Aldrich, Steinheim, Germany) and 2, 4-Pentanedione (Alfa Aesar, Karlsruhe, Germany) according to the reported paper [50]. The gel aging was allowed before drying and thermal treatment at 400 • C, in air, for 1 h. Subsequent hydrothermal treatment was applied in order to increase the defect concentration, as follows: the obtained powder was introduced in an autoclave containing hydrazine monohydrate (N 2 H 4 ×H 2 O 98%, Alfa Aesar) and aqueous solution of NaOH (Alfa Aesar) for 1 h at 140 • C.

Glutathione Capped Gold Nanoparticles
They were obtained by adapting the synthesis route reported by Wu et al. [51]. Gold (III) chloride trihydrate (HAuCl 4 ×3H 2 O ACS reagent > 49% Au) was dissolved in ultrapure water Millie-Q system, >18 MΩcm (24 mM) and kept at 0 • C under stirring. L-Glutathione reduced (GSH, > 98%, Carl Roth, 0.2 mmol) was dissolved in 100 mL ultrapure water Millie-Q system, >18 MΩcm, and slowly added to the gold solution. The clear mixture was kept under gentle stirring for 1 h. Sodium borohydride solution (NaBH 4 Sigma-Aldrich, 12 wt. % in 14 M NaOH) was added to the above-mentioned mixture that changes into a dark red solution that is added after 24 h to the annealed SiO 2 and TiO 2 powders. These are dried (30 min, 100 • C) and then thermally treated in the air for 3 h at 600 • C.

Glutathione Capped Platinum Nanoparticles
They were obtained by adapting the synthesis route reported by Eklund [52]. Chloroplatinic acid hexahydrate (H 2 PtCl 6 ×6H 2 O, Sigma-Aldrich) was dissolved in ultrapure water (12 mM) and subjected to a similar procedure to the previous one. Unlike that, the aqueous mixture GSH-H 2 PtCl 6 turns in yellow color when added NaBH 4 solution.

Modification of SiO 2 and TiO 2 with Ruthenium Dye
Ruthenizer (535-bis TBA from Solaronix, Aubonne, Switzerland) was used to obtain the sensitized oxides (RSiO 2 and RTiO 2 ) for increasing the generation of singlet oxygen under light irradiation. In this sense, Ruthenizer (0.0005 g in 0.5 mL ultrapure water) was added to 0.015 g annealed oxide powders (SiO 2 , TiO 2 ) and then dried at 60 • C.

PVA Hydrogel Synthesis
It follows the route reported by Hou et al. [11]. Briefly, 0.04 g Poly (vinyl alcohol) powder (Sigma-Aldrich, 80% hydrolyzed) was dissolved in 1 mL dimethyl sulfoxide (DMSO, Life Technologies, MW 78.13) and 0.5 mL H 2 O/D 2 O). In the case of gel samples exposed to visible light irradiation (for oxygen singlet checking), H 2 O is replaced by D 2 O. For UV-Vis characterization, 0.02 g oxide powder (bare and modified SiO 2 and TiO 2 ) is added to the freshly prepared gel aliquots (2 mL), and for antibacterial tests 0.001 g oxide powder is mixed with 10 µL gel sample. The resulting mixtures are gently shaken for 30 min.

Electron Microscopy Characterization
SEM images have been recorded with an FEI Quanta 3D microscope, FEG model. Morpho-structural characterization was carried out using JEM ARM200F Analytical Transmission Electron Microscope (JEOL Ltd., Tokyo, Japan).

AFM Characterization
Atomic force microscopy (AFM) measurements were performed with a microscope produced by Park Systems (XE-100 model) working in non-contact mode. The microscope allows obtaining accurate images due to its configuration, namely decoupled XY/Z, flexureguided cross talk eliminated scanners. In all AFM experiments sharp tips with less than 10 nm radius of curvature,~125 µm length,~30 µm width,~42N/m force constant, and 330 kHz resonance frequency (NCHR, Nanosensors™) were used. The AFM images were processed with the XEI program (v 1.8.0-Park Systems, Suwon, Korea) for thermal tilt correction and roughness assessment. The root mean squared roughness (Rq) represents the standard deviation of the height value in the image and the peak-to-valley parameter (Rpv) means the height difference between minimum and maximum. The AFM images are presented in classic mode (monochrome z-scale gradient) accompanied by representative line scans (plot of random line scans, each AFM image consisting of 256 lines) which show the surface profile in detail.

XRD and XRF Measurements
The Rigaku Ultima IV multipurpose diffraction system (Rigaku Corp., Tokyo, Japan) was used to obtain XRD patterns, employing a Cu-target tube (λ = 1.54060 Å). The working conditions included 30 mA and 40 kV, and the data were collected at room temperature within a 2θ range of 5 to 85 • , with a step size of 0.02 • and a scan rate of 2 • /min. For the elemental analysis of the materials (XRF measurement), the Rigaku ZSX Primus II spectrometer (Rigaku Corp., Tokyo, Japan) with wavelength dispersion in a vacuum was utilized. Rigaku's SQX analytical software was employed for semi-quantitative standardless analysis to compute the chemical composition of the samples.

XPS Characterization
XPS investigations were performed using a SPECS spectrometer with a PHOIBOS (150) analyzer equipped with non-monochromatized Mg Kα (1253.6 eV) X-rays anode radiation source operated at 300 W. Wide and detail spectra were registered at pressures lower than 2 × 10 −9 mbar, pass energy of 50 eV and 20 eV. The C1s peak was set at 285.00 eV for the binding energy scale. The spectra were fitted using Voigt peak profiles and a linear or a Shirley background (depending on shape), using the SDP v7.0 software (XPS International, Salem, Oregon, United States).

Diffuse Reflectance UV-Vis Characterization
Diffuse reflectance UV-Vis spectra were obtained using a spectrophotometer Perkin Elmer Lambda 35, equipped with an integrating sphere. The measurements were carried out in the range of 1100-250 nm, using Spectralon as a reference. The reflectance measurements were converted to absorption spectra using the Kubelka-Munk function, F(R). For determining the UV-Vis light absorption properties of the gels, the scattered transmission was measured using the same instrument (Perkin Elmer, Lambda 35). The sample was placed in micro-cuvettes (Brand: 12.5 mm × 12.5 mm × 45 mm, center height: 8.5 mm, 230-900 nm, Roth) and introduced in the transmission port of the integrating sphere. The transmitted and forward scattered light was collected and measured by the sphere (during the measurements the Spectralon was placed in the reflectance port of the instrument). Target samples (0.001 g powder or 0.5 mL gel) were suspended in methanolic solution of SOSG (Singlet Oxygen Sensor Green-Thermo Fisher Scientific/Invitrogen) contained by quartz cuvettes and exposed to visible irradiation (provided by a Peccel solar simulator equipped with cut off filter λ > 420 nm Asahi Spectra). The photogenerated oxygen singlet reacts with a SOSG component leading to the endoperoxide formation signaled by the appearance of the photoluminescence peak around 530 nm [35,53,54] (λexc = 480 nm, excitation and emission slits 2.5/2.5). This is depicted with a Carry Eclipse fluorescence spectrometer, Agilent Technologies.

(ii) Hydroxyl radicals (•OH) generation under simulated solar light
The samples of interest (0.001 g powder or 0.5 mL gel) were suspended in a coumarin solution (10 mM, Merck) and exposed to simulated solar irradiation (Peccel Solar Simulator, Tokyo, Japan). The photogenerated hydroxyl radicals react with coumarin leading to the formation of a fluorescent product, umbelliferone, that is evaluated according to a characteristic PL emission peak from 450 nm for λexc = 330 nm [55,56] (excitation and emission slits 5/10, Carry Eclipse fluorescence spectrometer).

Electrokinetic Potential Measurements
Electrokinetic potential measurements have been performed on a Malvern Nano ZS Zetasizer, Model ZEN 3600 at room temperature. The measurements were performed in triplicate, the values being averaged.

Antibacterial Activity
The antimicrobial activity of the investigated samples (powders and gels containing powders) was assessed against the bacterial strain Staphylococcus (S.aureus ATCC 29213). The bacterial strain was grown in LB liquid medium (g/L: peptone 10 g, NaCl 5 g, yeast extract 5 g, pH 6) together with the samples of interest (under light exposure or in dark, at 37 • C).
Firstly, the cell growth in the dark was evaluated by measuring the optical density (OD) at 600 nm with the multireader Bio Tek Synergy HTX, Agilent US. All tests were performed in triplicate.
To highlight the dependence of the bacterial cells' viability after 24 h on the sample characteristics and testing conditions (type of irradiation, powder amount), the decimal dilutions assay was performed. The successive decimal dilutions (1:10 v/v) are necessary to obtain a less dense suspension that is scattered on solid nutritional broth. After incubation on optimal conditions (37 • C, 18-24 h) the number of isolated colonies is counted, and the UFC/mL value is determined by multiplying with the dilution factor (Ben-David A, 2014). S. aureus ATCC 29213 grown in LB liquid broth was used as control. For samples exposed to visible light irradiation, the bacterial cell culture contained D 2 O instead of H 2 O and for solar irradiation contained D 2 O and H 2 O.
Supplementary Materials: The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/gels9080650/s1, Figure S1: Topographic 2D AFM image of the TiO 2 powder modified with AuNPs; Figure S2: Part (slice) of a SiO 2 tube exhibited in a 3D topographic AFM image o at the scale of 4 µm × 25 µm; Figure S3: Part of a SiO 2 tube exhibited in a 3D topographic AFM image o at the scale of 6 µm × 8 µm; Figure S4: The generation of the hydroxyl radicals under simulated solar irradiation by SiO 2 and TiO 2 based powders, free and embedded in PVA gel; Figure S5: Electrokinetic potential measurements for SiO 2 and TiO 2 -based powders free and embedded in PVA gel. The experiments were made in triplicates.