Self-Cleaning Coatings for the Protection of Cementitious Materials: The Effect of Carbon Dot Content on the Enhancement of Catalytic Activity of TiO 2

: The urgent demand for pollution protection of monuments and buildings forced the interest towards speciﬁc preservation methods, such as the application of photocatalytic coatings with self-cleaning and protective activity. TiO 2 photocatalysts without and with a variety of carbon dots loading (TC0, TC25–75) were synthesized via a green, simple, low cost and large-scale hydrothermal method using citric acid, hydroxylamine and titanium isopropoxide (TTIP) and resulted in uniform anatase phase structures. In photocatalysis experiments, TC25 and TC50 composites with 1:3 and 1:1 mass ratio of C-dots solution to TTIP, respectively, showed the best degradation efﬁciency for methyl orange (MO) under UV-A light, simulated solar light and sunlight compared to TiO 2 , commercial Au/TiO 2 (TAu) and catalysts with higher C-dot loading (TC62.5 and TC75). Treatment of cement mortars with a mixture of photocatalyst and a consolidant (FX-C) provided self-cleaning activity under UV-A and visible light. This study produced a variety of new, durable, heavy metal-free C-dots/TiO 2 photocatalysts that operate well under outdoor weather conditions, evidencing the C-dot dosage-dependent performance. For the building protection against pollution, nanostructured photocatalytic ﬁlms were proposed with consolidation and self-cleaning ability under solar irradiation, deriving from combined protective silica-based agents and TiO 2 photocatalysts free or with low C-dot content.


Introduction
Air pollution, the fourth greatest fatal health risk so far, is accounting for serious danger to human health, causing millions of premature deaths annually [1]. There is a demanding need to combat climate change as far as the global population grows; thus, highly efficient technologies can serve as the key to improving air quality and protecting the environment [2,3]. Moreover, air pollution and climate change cause physical and aesthetic damage to buildings and cultural heritage constructions, accelerating the decay process [4]. For the protective treatment of several types of building substrates, advanced coatings with self-cleaning ability have been developed throughout the years [5]. The main advantage of photocatalytic air and wastewater purification is that chemicals or external energy input are not requisite, except for ambient light or sunlight, which is not costly. Despite the strong effort for the development of various novel active materials, titanium dioxide (TiO 2 ) is by far the most promising and widely employed technology in photocatalysis among them [6].
TiO 2 was introduced in the field of photocatalysis in 1972 [7], and since then, it has become the most investigated compound for air purification due to its excellent physicochemical stability and oxidation properties, as well as cost-effectiveness, high availability and eco-friendliness [8]. Photoinduced processes begin with the absorption of electromag-dots nanocomposites were used as model photocatalysts to investigate the effect of the C-dot content in photocatalytic degradation of MO under UV-A and visible light, but also direct sunlight radiation, showing remarkable results compared to pure TiO 2 (TC0), as well as commercially available Au/TiO 2 (TAu) and this is the first innovative result of this study. The most suitable photocatalysts were found to be TC25 and TC50, evidencing that this range of C-dot loading is superior for the decomposition of organic pollutants, another interesting finding that correlates Cdots amount and efficiency. Another innovation of this study, apart from the rapid and excellent photocatalytic activity of these catalysts under solar light irradiation, is the almost eternal nature, as they can be recycled and reused up to ten times without losing their catalytic ability (~90% degradation after 3 h of irradiation into the 10th cycle, >99% after 6 h).
Furthermore, based on the tetraethoxysilane (TEOS) effectiveness as a strengthening agent for Portland cement mortars [28], we synthesized an advanced hydrophobic consolidant, designated as FX-C, using a sol-gel process that combines TEOS, PDMS and nano-calcium oxalate (nano-CaOx) for the preservation of cultural heritage [29,30]. It was found that silicon in TEOS/PDMS copolymer can form strong Si-O bonds with hydroxyl groups of the material for preservation [31]. It is well-established that among other useful properties of TiO 2 -based catalysts [32][33][34], coating application of nano-TiO 2 or SiO 2 @TiO 2 [35][36][37][38] in mortars provides restoration and self-cleaning ability in building materials [39][40][41], while TEOS-PDMS-TiO 2 hybrid nanomaterials in limestone and marbles enhance these properties [42,43]. Our previous work that studied the self-cleaning ability of the composite coating FX-C with TiO 2 /C-dots in concrete, limestones and lime mortars showed a moderate photocatalytic activity of this protective agent under visible light irradiation [26]. The fourth innovative result of this work refers to the accomplishment of a self-cleaning coating of FX-C and TC0-TC25 activated under visible light irradiation, ready to be applied onto cementitious mortars.

Materials
The citric acid (monohydrate, ≥99%) and hydroxylamine hydrochloride (99%) were employed from Sigma-Aldrich (St. Louis, MI, USA) as C-dots precursors and used as received. Titanium (IV) isopropoxide (TTIP, ≥97%) and hydrochloric acid (~37%) were purchased from Sigma-Aldrich (St Louis, MI, USA) and used as received for the synthesis of TC0-75 powders. Methyl Orange (for microscopy, Hist.) was obtained from Sigma Aldrich (St Louis, MI, USA) and utilized as the model compound for the degradation of organic pollutants. Gold 1% on titanium dioxide extrudates was employed from Strem Chemicals (Newburyport, MA, USA) and used as received. Commercially available solvents were used as received without further purification.

Analytical Techniques
High-resolution TEM micrographs of C-dots and TC0-75 were obtained using a JEM-2100 transmission electron microscope and a JEM-100C microscope (JEOL Ltd., Tokyo, Japan), operating at an accelerating voltage of 200 kV and 80 kV, respectively. The samples for TEM were prepared by drop-casting 8 µL of the diluted aqueous solution of each sample onto Formvar/carbon-copper-coated grids supplied by Agar Scientific (Essex, UK) and then left to dry at room temperature for 2 h. A transmission electron microscope was also used for statistical size-distribution histograms by counting and measuring the diameter size distribution via ImageJ software. SEM images were recorded on a JSM 6390LV scanning electron microscope (JEOL Ltd., Tokyo, Japan) operated at 20 kV electron voltage. TC nanoparticles were dried on a two-sided carbon tape and sputter-coated with ca. 100 Å of gold (Au), using an SCD 050 sputter coater (BAL-TEC, Los Angeles, CA, USA). UV-Vis spectra of samples were obtained on a Cary 1E UV-Vis spectrophotometer (Varian, Palo Alto, CA, USA), in the wavelength range between 200 and 700 nm, using a quartz cuvette with 10 mm path length and 3.5 mL chamber volume. UV-Vis/near-IR diffuse reflectance spectra were recorded on a UV-2600 spectrophotometer (Shimadzu, Kyoto, Japan) using BaSO 4 as the 100% reflectance reference. Powder X-ray diffraction data (XRD) of the TC0-75 catalysts were obtained from a D8 Advance X-ray diffractometer (Bruker, Billerica, MA, USA) equipped with a Lynx Eye strip silicon detector (Bruker, Billerica, MA, USA). Measurements were applied at room temperature, using Cu Ka radiation at 35 mA and 35 kV. Fourier Transform Infrared spectroscopy (FT-IR) in powdered samples was performed with an iS50 FT-IR spectrometer (Thermo Fischer Scientific, Waltham, MA, USA) in the spectral range between 4000 and 400 cm −1 . For Attenuated Total Reflection (ATR) measurements, a built-in, all-reflective diamond was used as the internal reflection element. The measurement of the color change via photocatalytic degradation was performed with a CM-2600d spectrophotometer (Konica Minolta, Tokyo, Japan) with a D65 illuminant at 8-degree viewing, in the wavelength range between 360 and 740 nm. A 36 W UV EBN001W LED curing lamp (340-380 nm, maximum emission at 365 nm, Esperanza sp. j. Poterek, Ożarów Mazowiecki, Poland) provided the UV-A irradiation source, while a laboratory constructed stimulated solar box equipped with two L15W/840 tubular fluorescent lamps (400-630 nm, maximum emission at 540 nm, OSRAM, Munich, Germany) was used as the solar simulator. A GM3120 (BENETECH, Shenzhen, China) electromagnetic radiation tester was used to measure light irradiance.

Synthesis of Carbon Dots
Regarding the C-dots preparation, the procedure was slightly modified from an already published process [44]. A solution of citric acid (1.65 g) and hydroxylamine hydrochloride (1.65 g) in deionized water (16.5 mL) was heated at 300 • C for 12 min. The resulting dark brown, sticky solid was dissolved in water, sonicated via ultrasonic agitation for 10 min to separate the aggregates and filtered through a 7-12 µm Whatman filter paper to remove undissolved byproducts. A final 2.5% w/v C-dots solution was collected as an opaque liquid ( Figure 1). UV-Vis spectra of samples were obtained on a Cary 1E UV-Vis spectrophotometer (Varian, Palo Alto, CA, USA), in the wavelength range between 200 and 700 nm, using a quartz cuvette with 10 mm path length and 3.5 mL chamber volume. UV-Vis/near-IR diffuse reflectance spectra were recorded on a UV-2600 spectrophotometer (Shimadzu, Kyoto, Japan) using BaSO4 as the 100% reflectance reference. Powder X-ray diffraction data (XRD) of the TC0-75 catalysts were obtained from a D8 Advance X-ray diffractometer (

Synthesis of Carbon Dots
Regarding the C-dots preparation, the procedure was slightly modified from an already published process [44]. A solution of citric acid (1.65 g) and hydroxylamine hydrochloride (1.65 g) in deionized water (16.5 mL) was heated at 300 °C for 12 min. The resulting dark brown, sticky solid was dissolved in water, sonicated via ultrasonic agitation for 10 min to separate the aggregates and filtered through a 7-12 μm Whatman filter paper to remove undissolved byproducts. A final 2.5% w/v C-dots solution was collected as an opaque liquid ( Figure 1).

Synthesis of TiO2 (TC0) and TC25-75 Photocatalysts
For the preparation of TiO2 with different loading of C-dots, an already published method was followed [26], employing slight modifications ( Figure 2). Briefly, a 2.5% w/v aqueous C-dots solution, specifically 1.5 g for TC25 up to 13.5 g for TC75, was dissolved in a 1:1 mixture of ethanol and deionized water (50 mL). For this solution, a solution of TTIP (4.5 g) in ethanol (20 mL) was added dropwise, and the pH of the resulting mixture was adjusted to 2.5 by the addition of concentrated HCl (~5 drops) to facilitate the hydrolysis of TTIP. The mixture was allowed to stir at ambient temperature overnight, followed by heating at 80 °C for 8 h to facilitate titania polymerization and C-dots incorporation. Afterward, the mixture was cooled to ambient temperature, centrifuged for 10 min at 3000 rpm, and the wet solid was dried at 80 °C and ground using a mortar. Finally, the catalyst was calcinated at 200 °C for 2 h to produce the crystalline material as a yellow (TC25) to brown solid (TC50-75) in 85%-98% yield.

Synthesis of TiO 2 (TC0) and TC25-75 Photocatalysts
For the preparation of TiO 2 with different loading of C-dots, an already published method was followed [26], employing slight modifications ( Figure 2). Briefly, a 2.5% w/v aqueous C-dots solution, specifically 1.5 g for TC25 up to 13.5 g for TC75, was dissolved in a 1:1 mixture of ethanol and deionized water (50 mL). For this solution, a solution of TTIP (4.5 g) in ethanol (20 mL) was added dropwise, and the pH of the resulting mixture was adjusted to 2.5 by the addition of concentrated HCl (~5 drops) to facilitate the hydrolysis of TTIP. The mixture was allowed to stir at ambient temperature overnight, followed by heating at 80 • C for 8 h to facilitate titania polymerization and C-dots incorporation. Afterward, the mixture was cooled to ambient temperature, centrifuged for 10 min at 3000 rpm, and the wet solid was dried at 80 • C and ground using a mortar. Finally, the catalyst was calcinated at 200 • C for 2 h to produce the crystalline material as a yellow (TC25) to brown solid (TC50-75) in 85%-98% yield. The same process, in the absence of the aqueous C-dots solution, was applied for the synthesis of TiO2 as a control substance ( Figure 2). It is worth mentioning that a crucial factor in this experiment is the adjustment of pH at 2.5 since lower pH leads to a reduction in the isolated yield of the catalyst. For example, at pH 1.5, the isolated yield of TC50 drops to 46%. Furthermore, in the process of synthesis of the control substance TC0, acidification up to pH 1.5 leads to a mixture of brookite and anatase phases of TiO2, whereas at pH 2.5, only the anatase phase is present, as indicated in the XRD analysis.

Incorporation of Photocatalysts into the Protective Agent
Consolidant FX-C was prepared in the laboratory by an already published method [29]. Each solid catalyst (4% w/w) was added to the sol FX-C, and the mixture was sonicated via ultrasonic agitation for 10 min to ensure homogeneity ( Figure S1).

Application of the Protective/Photocatalytic Film onto Cementitious Mortars
The performance of the photocatalytic films in the conservation of cement-based materials was evaluated using cement mortars prepared in the laboratory, with a binder/sand ratio of 1:3, based on local 0-3 mm of grain size standard sand of carbonaceous nature. Further information on the mix design of the treated mortars can be found in Table S1. The synthesized photocatalytic films were brushed three times on the cement mortar substrates. Control experiments were also performed by measuring the untreated cement mortar and by brushing the cementitious material with FX-C in the absence of a photocatalyst.

Photodegradation of MO Using Photocatalysts and Their Reusability
The activity of a series of TC nanocomposites on the photodegradation of MO was initially evaluated under UV-A light, using a 36 W LED curing lamp (maximum emission at 365 nm) with a light irradiance of approximately 33.6 mW/cm 2 and artificial solar light, using a solar box equipped with two 15 W tubular fluorescent lamps (maximum emission at 540 nm) with a light irradiance of approximately 13.6 mW/cm 2 . For both irradiation types, the distance of each sample from the lamps was set to 7 cm.
The photocatalytic activity of TC25, TC50, TC62.5 and TC75 catalysts, compared to TC0, was evaluated by studying the photodegradation of MO solution in the presence of the appropriate catalyst under UV-A light, simulated solar light, as well as sunlight in outdoor weather conditions. For this purpose, the solid photocatalyst (40 mg, 0.08% w/v) was dispersed to a 5 ppm MO aqueous solution (50 mL), and the mixture was allowed to stir in darkness for 30 min in order to reach adsorption-desorption equilibrium of dye on The same process, in the absence of the aqueous C-dots solution, was applied for the synthesis of TiO 2 as a control substance ( Figure 2). It is worth mentioning that a crucial factor in this experiment is the adjustment of pH at 2.5 since lower pH leads to a reduction in the isolated yield of the catalyst. For example, at pH 1.5, the isolated yield of TC50 drops to 46%. Furthermore, in the process of synthesis of the control substance TC0, acidification up to pH 1.5 leads to a mixture of brookite and anatase phases of TiO 2 , whereas at pH 2.5, only the anatase phase is present, as indicated in the XRD analysis.

Incorporation of Photocatalysts into the Protective Agent
Consolidant FX-C was prepared in the laboratory by an already published method [29]. Each solid catalyst (4% w/w) was added to the sol FX-C, and the mixture was sonicated via ultrasonic agitation for 10 min to ensure homogeneity ( Figure S1).

Application of the Protective/Photocatalytic Film onto Cementitious Mortars
The performance of the photocatalytic films in the conservation of cement-based materials was evaluated using cement mortars prepared in the laboratory, with a binder/sand ratio of 1:3, based on local 0-3 mm of grain size standard sand of carbonaceous nature. Further information on the mix design of the treated mortars can be found in Table S1. The synthesized photocatalytic films were brushed three times on the cement mortar substrates. Control experiments were also performed by measuring the untreated cement mortar and by brushing the cementitious material with FX-C in the absence of a photocatalyst.

Photodegradation of MO Using Photocatalysts and Their Reusability
The activity of a series of TC nanocomposites on the photodegradation of MO was initially evaluated under UV-A light, using a 36 W LED curing lamp (maximum emission at 365 nm) with a light irradiance of approximately 33.6 mW/cm 2 and artificial solar light, using a solar box equipped with two 15 W tubular fluorescent lamps (maximum emission at 540 nm) with a light irradiance of approximately 13.6 mW/cm 2 . For both irradiation types, the distance of each sample from the lamps was set to 7 cm.
The photocatalytic activity of TC25, TC50, TC62.5 and TC75 catalysts, compared to TC0, was evaluated by studying the photodegradation of MO solution in the presence of the appropriate catalyst under UV-A light, simulated solar light, as well as sunlight in outdoor weather conditions. For this purpose, the solid photocatalyst (40 mg, 0.08% w/v) was dispersed to a 5 ppm MO aqueous solution (50 mL), and the mixture was allowed to stir in darkness for 30 min in order to reach adsorption-desorption equilibrium of dye on the catalyst surface. Thereafter, the mixture was irradiated with the appropriate light source for a period of 120 min, with a sample removal (4 mL) every 15 min for reaction monitoring. Analytical determination of the reduction in MO concentration during the Coatings 2022, 12, 587 6 of 18 photocatalytic degradation process was carried out spectrophotometrically at 465 nm. A commercially available catalyst Au (1%)/TiO 2 (TAu) was also used in the MO degradation under sunlight conditions for comparison. Under the same conditions, MO degradation was tested in the absence of a catalyst, where no decomposition was observed.
Furthermore, the reusability of TC50 was tested under simulated sunlight irradiation. In this case, TC50 (0.16 g) was added into a 5 ppm aqueous MO solution (80 mL) and irradiated with visible light for 180 min. After each cycle, the solution was decanted, and the catalyst was washed with water and ethanol (30 mL each), dried at 60 • C overnight, and then reused in the next cycle. This process was repeated 9 times, while at the end of the 10th cycle, the catalyst was recycled (0.08 g) and could be further used in the next experiment.

Self-Cleaning Performance of Treated Cementitious Mortars
In order to determine the self-cleaning and air depolluting capabilities of the protective films enhanced with TiO 2 /C-dots, the protective coating was brushed onto the cement mortar substrates, and the photodegradation of a drop of MB under UV-A and visible light was determined using a portable CM-2600d spectrophotometer (Konica Minolta, Tokyo, Japan). Control experiments were performed by exposing the untreated cement mortar and the cementitious material with FX-C, covered with the same amount of MB. The total color difference (∆E) can be calculated using the following equation [45]: where the differences in color (∆) with time, for the treated with a protective coating and spotted with MB cement mortars, were expressed as L*, a* and b* values of the CIELab color space. The L* values range from 0 for black to +100 for white, while the negative and positive a* values represent green and red, respectively. As for the negative and positive b* values, these represent blue and yellow, respectively. the catalyst surface. Thereafter, the mixture was irradiated with the appropriate light source for a period of 120 min, with a sample removal (4 mL) every 15 min for reaction monitoring. Analytical determination of the reduction in MO concentration during the photocatalytic degradation process was carried out spectrophotometrically at 465 nm. A commercially available catalyst Au (1%)/TiO2 (TAu) was also used in the MO degradation under sunlight conditions for comparison. Under the same conditions, MO degradation was tested in the absence of a catalyst, where no decomposition was observed.

Structure and Morphology of C-Dots and TC0-75 Photocatalysts
Furthermore, the reusability of TC50 was tested under simulated sunlight irradiation. In this case, TC50 (0.16 g) was added into a 5 ppm aqueous MO solution (80 mL) and irradiated with visible light for 180 min. After each cycle, the solution was decanted, and the catalyst was washed with water and ethanol (30 mL each), dried at 60 °C overnight, and then reused in the next cycle. This process was repeated 9 times, while at the end of the 10th cycle, the catalyst was recycled (0.08 g) and could be further used in the next experiment.

Self-Cleaning Performance of Treated Cementitious Mortars
In order to determine the self-cleaning and air depolluting capabilities of the protective films enhanced with TiO2/C-dots, the protective coating was brushed onto the cement mortar substrates, and the photodegradation of a drop of MB under UV-A and visible light was determined using a portable CM-2600d spectrophotometer (Konica Minolta, Tokyo, Japan). Control experiments were performed by exposing the untreated cement mortar and the cementitious material with FX-C, covered with the same amount of MB. The total color difference (ΔE) can be calculated using the following equation [45]: where the differences in color (Δ) with time, for the treated with a protective coating and spotted with MB cement mortars, were expressed as L*, a* and b* values of the CIELab color space. The L* values range from 0 for black to +100 for white, while the negative and positive a* values represent green and red, respectively. As for the negative and positive b* values, these represent blue and yellow, respectively.   as rutile or brookite. The anatase crystalline phase is commonly accepted as a more efficient photocatalyst, compared to rutile and brookite, due to its high photocatalytic activity [46]. Crystallite average size of the TC0-75 nanoparticles was estimated using the following Scherrer equation [47]:

Structure and Morphology of C-Dots and TC0-75 Photocatalysts
where D is the crystallite particle size (in nm), K is the Scherrer constant equal to~0.9 in the case of spherical shapes, λ is the wavelength of the X-ray's radiation (0.15406 nm), β is the full width at half maximum (FWHM) height of the diffraction peaks and θ is the Bragg's diffraction angle.  [48][49][50], as well as the stability of anatase when the particle size is below 14 nm [51]. TEM images of the as-synthesized C-dots ( Figure S3) were similar to that observed in our previous work [26]. The average size of C-dots was 1.9 nm, ranging from 1.2 to 2.6 nm, and the shape appeared to be uniform and quasi-spherical ( Figure S3c). TEM analysis of the TC50 and TC75 photocatalysts can be seen in Figure 4a,b. TC50 exhibited relatively uniform size nanoparticles, with an average size of 20-21 nm and the size distribution ranging approximately from 14 to 32 nm. In a previous work, a similar size distribution ranging from 10 to 40 nm for the TC25 photocatalyst was observed [26]. Agglomeration of crystallites resulting in nanocluster formation was observed in previous works for TiO 2 [52,53] and TiO 2 doped with C-dots [10] or with other species [54,55]. TEM images of TC75 nanoparticles in Figure 4c,d show a non-uniform but quasi-spherical shape in an average size of 64-68 nm and wide size distribution of 40 to 100 nm ( Figure S4) due to the high C-dot loading in the TiO 2 surface. The XRD results and the crystal planes observed in the TEM images of our previous work for TC25 [26] were both attributed to the crystalline structure of the TiO 2 /C-dots nanoparticles, further evidenced in the SEM images in Figure 4. Even though the synthesis of pure anatase requires calcinations at 500 • C within several hours [56], nevertheless, in our case, TEM images indicated that at a relatively low temperature, a C-dots-assisted crystalline TiO 2 could be formed [26].
Indeed, SEM was used to display the surface morphology of TC25 and TC62.5 nanocomposites ( Figure 5). Both nanohybrids consist of nearly spherical in shape, rough and compact aggregates with crystal structures and various diameters. The spherical-like shape is typical for anatase crystallite structure [10].
The mineralogical and microstructural characterization of the studied photocatalysts revealed that anatase was formed in the presence of different C-dot loading upon a heating process up to 200 • C. By increasing the C-dot loading, as in the case of the TC75, the average size of the photocatalysts was also increased due to agglomeration phenomena.
The FTIR spectrum of C-dots ( Figure 6) confirms the presence of polar groups derived from the incorporation of hydroxylamine into the carbon core. This observation is in accordance with the good dispersion properties that C-dots exhibit in water. More specifically, the broad peaks at 3450 and 3200 cm −1 correspond to the symmetric stretching vibration of O-H and N-H, respectively [23], while the peaks at 1605 and 1399 cm −1 can be attributed to the asymmetric and symmetric stretching vibration of COO − [57]. The stretching vibration of C=O at 1697 cm −1 along with the bending vibration of N-H in amides at 1584 cm −1 [26,58] and the stretching vibration of C-O in carboxylates at 1183 cm −1 are also present in the C-dots spectrum. Other characteristic peaks of the main core in carbon dots can be considered: (a) the asymmetric and symmetric stretching vibration of C-H bonds with the broad peaks at 3041 and 2807 cm −1 , and (b) the stretching and bending vibrations of C=C at 1630 and 746 cm −1 . Finally, it is important to highlight that in the FTIR spectra of the TC25-75 photocatalysts (Figure 6), characteristic bands of C-dots at 1698, 1605 and 1183 cm −1 were shifted to higher and lower wavenumbers (e.g., 1630, 1552 and 1223 cm −1 , respectively), due to the C-dots' incorporation in the TiO 2 surface. In the sample TC75, the higher intensity of the C-H asymmetric and symmetric stretching vibrations at 2921 and 2852 cm −1 is directly correlated with the higher C-dots content.  Indeed, SEM was used to display the surface morphology of TC25 and TC62.5 nanocomposites ( Figure 5). Both nanohybrids consist of nearly spherical in shape, rough and compact aggregates with crystal structures and various diameters. The spherical-like shape is typical for anatase crystallite structure [10].   Indeed, SEM was used to display the surface morphology of TC25 and TC62.5 nanocomposites ( Figure 5). Both nanohybrids consist of nearly spherical in shape, rough and compact aggregates with crystal structures and various diameters. The spherical-like shape is typical for anatase crystallite structure [10]. The mineralogical and microstructural characterization of the studied photocatalysts revealed that anatase was formed in the presence of different C-dot loading upon a heating process up to 200 °C. By increasing the C-dot loading, as in the case of the TC75, the average size of the photocatalysts was also increased due to agglomeration phenomena.
The FTIR spectrum of C-dots ( Figure 6) confirms the presence of polar groups derived from the incorporation of hydroxylamine into the carbon core. This observation is in accordance with the good dispersion properties that C-dots exhibit in water. More specifically, the broad peaks at 3450 and 3200 cm −1 correspond to the symmetric stretching vibration of O-H and N-H, respectively [23], while the peaks at 1605 and 1399 cm −1 can be attributed to the asymmetric and symmetric stretching vibration of COO − [57]. The stretching vibration of C=O at 1697 cm −1 along with the bending vibration of N-H in amides at 1584 cm −1 [26,58] and the stretching vibration of C-O in carboxylates at 1183 cm −1 are also present in the C-dots spectrum. Other characteristic peaks of the main core in carbon dots can be considered: (a) the asymmetric and symmetric stretching vibration of C-H bonds with the broad peaks at 3041 and 2807 cm −1 , and (b) the stretching and bending vibrations of C=C at 1630 and 746 cm −1 . Finally, it is important to highlight that in the FTIR spectra of the TC25-75 photocatalysts ( Figure 6), characteristic bands of C-dots at 1698, 1605 and 1183 cm −1 were shifted to higher and lower wavenumbers (e.g., 1630, 1552 and 1223 cm −1 , respectively), due to the C-dots' incorporation in the TiO2 surface. In the sample TC75, the higher intensity of the C-H asymmetric and symmetric stretching vibrations at 2921 and 2852 cm −1 is directly correlated with the higher C-dots content. The optical properties of the TC0-75 photocatalysts were studied by UV-Vis/NIR diffuse reflectance spectroscopic (DRS) measurement (Figure 7a). It is obvious that the absorption edge of TC25-75 was shifted towards the visible light region compared to TC0, suggesting significant absorption ability for solar light. The energy band gaps (Eg) of photocatalysts TC0-75 were determined from UV-Vis/NIR diffuse reflectance spectra ( Figure  7b), using Tauc plot analysis for indirect allowed transition [59], due to the anatase crystallite form of TC structures. In this case, (f hv) 1/2 is expressed as a function of photon energy (hv), where f is the Kubelka-Munk function [60] of the measured reflectance (R), as shown in the following equation: The optical properties of the TC0-75 photocatalysts were studied by UV-Vis/NIR diffuse reflectance spectroscopic (DRS) measurement (Figure 7a). It is obvious that the absorption edge of TC25-75 was shifted towards the visible light region compared to TC0, suggesting significant absorption ability for solar light. The energy band gaps (Eg) of photocatalysts TC0-75 were determined from UV-Vis/NIR diffuse reflectance spectra (Figure 7b), using Tauc plot analysis for indirect allowed transition [59], due to the anatase crystallite form of TC structures. In this case, (f hv) 1/2 is expressed as a function of photon energy (hv), where f is the Kubelka-Munk function [60] of the measured reflectance (R), as shown in the following equation: 7b), using Tauc plot analysis for indirect allowed transition [59], due to the anatase crystallite form of TC structures. In this case, (f hv) 1/2 is expressed as a function of photon energy (hv), where f is the Kubelka-Munk function [60] of the measured reflectance (R), as shown in the following equation:  This analysis led to Eg values of 3.10 eV for TC0, which is close to the bandgap of anatase. On the contrary, the bandgaps of TC25 and TC50 were measured at 2.97 eV each, while TC75 exhibited a bandgap of 3.18 eV. The findings are in correlation with the photocatalysis results, suggesting that the Eg of the nanocomposites appears to be an important parameter that accounts for the efficiency of the photocatalytic process. Thus, C-dots can reduce the energy bandgap of TiO 2 up to a point where C-dots dominate the structure of the photocatalyst, leading to an even higher Eg than TiO 2 , as in the case of TC75 with higher C-dot content.

Photocatalytic Performance of TC25-75 Compared to TCO and TAu
The results of the UV-A irradiation process were shown in Figure 8a, where TC25-62.5 composites provided a much higher photodegradation rate for MO compared to TiO 2 (TC0) and TC75, with both TC25 and TC50 photocatalysts showing the best performance. The light irradiance of the 36 W LED curing lamp used in our experiments, based on the sample distance, was approximately 33.6 mW/cm 2 . The area of a 50 mL beaker that was used in the experiments can be defined as half the curved surface area of the cylinder, πrh, where r is the base radius, and h is the height of the cylinder. Therefore, the average continual radiation was~1.42 W. After 60 min of UV-A light irradiation, the MO degradation rates in the presence of TC0, TC25, TC50, TC62.5 and TC75 were 20.8, 91.4, 92.0, 56.9 and 48.7%, respectively. C-dots had almost no photocatalytic ability for MO, as only a 4.6% of the dye was degraded after 60 min. The UV-Vis spectrum and macroscopic evaluation of MO degradation using TC50 under UV-A light are displayed in Figures S5 and S6. Similar results, in a slower degradation rate, were found during the simulated solar light irradiation, as shown in Figure 8b. The light irradiance of the two 15 W fluorescent lamps, based on the sample distance, was measured at 13.6 mW/cm 2 ; thus, the average continual radiation was 0.58 W. It is evident that all of the TC25-75 composites exhibited higher photocatalytic activity for MO compared to TC0, whereas both TC25 and TC50 provided the highest degradation rate. More specifically, after 120 min of irradiation, the MO degradation rates in the presence of TC0, TC25, TC50, TC62 and TC75 were 45.4, 96.2, 90.9, 62.9 and 46.1%, respectively. The UV-Vis spectrum and macroscopic evaluation of MO degradation using TC50 under visible light are reported in Figures S7 and S8. tocatalytic activity for MO compared to TC0, whereas both TC25 and TC50 provided the highest degradation rate. More specifically, after 120 min of irradiation, the MO degradation rates in the presence of TC0, TC25, TC50, TC62 and TC75 were 45.4, 96.2, 90.9, 62.9 and 46.1%, respectively. The UV-Vis spectrum and macroscopic evaluation of MO degradation using TC50 under visible light are reported in Figures S7 and S8.  In all photocatalytic experiments, the concentration of MO was fixed at 5 ppm. It is worth mentioning that for concentrations ranging from 5 to 25 ppm, the highest photocatalytic activity was found to be 20 ppm in an anatase titania suspension system, with 5 ppm being the least suitable MO concentration [61]. MO exhibits the lowest degradation rate compared to RhB and MB because an electron is more difficult to be transferred to TiO 2 , as the Highest Occupied Molecular Orbital (HOMO) energy gap of MO is −5.624 eV, compared to −10.128 eV for RhB and −10.494 for MB [62]. For those reasons, the high photocatalytic activity of nanocomposites TC25 and TC50 in MO degradation can be described as a remarkable result. Furthermore, the same photocatalytic group of TC nanocomposites was tested on the photodegradation of MO under solar light in outdoor weather conditions. According to accurate weather station data [63], during the days that the experiments took place in our department area, the daily all-sky surface shortwave downward irradiance was in the range of 2.46-7.49 KWh/m 2 /day (102.5-312.1 W/m 2 ). Thus, in our process, the average continual radiation was in the range of 0.43-1.32 W, and the degradation rates can be summarized in Figure 9. In this case, in the presence of the photocatalysts, TC25 and TC50 MO rapidly becomes degraded, as the rate increases to 96.0% and 90.2% for the first 30 min of irradiation ( Figures S9 and S10). It must be pointed out that control experiments performed under the conditions of irradiation in the absence of the photocatalyst did not show any change in the azo-dye solution absorbance. It can also be noted that under similar reaction conditions, a lower degradation rate (similar to TC75) was obtained with a commercially available gold catalyst, supported by TiO 2 (TAu).
The results demonstrate that in all cases, the photocatalytic performance of the composite materials strongly depends on the C-dots percentage in TiO 2 . A remarkable photocatalytic result in the case of low to moderate C-dot loading turns to reduced activity when the C-dot loading increases in the TiO 2 core. This could be attributed to the excess of C-dots on the surface of TiO 2 in TC75, which cause blockage of the pore channels, thus hindering light-harvesting and charge transfer [23,64]. commercially available gold catalyst, supported by TiO2 (TAu).
The results demonstrate that in all cases, the photocatalytic performance of the composite materials strongly depends on the C-dots percentage in TiO2. A remarkable photocatalytic result in the case of low to moderate C-dot loading turns to reduced activity when the C-dot loading increases in the TiO2 core. This could be attributed to the excess of Cdots on the surface of TiO2 in TC75, which cause blockage of the pore channels, thus hindering light-harvesting and charge transfer [23,64]. Based on the heterogeneity of the catalytic system, the TC catalysts can be easily separated from the reaction mixture by careful decantation of the aqueous solution or, more accurately, via centrifugation, followed by washing with water and ethanol. Thus, the stability of the TC50 catalyst was examined by performing ten consecutive catalytic tests by irradiating 5 ppm of MO in the presence of 0.2% w/v photocatalyst for 180 min under simulated solar light. As it can be seen in Figure 10, even after 10 cycles, 89.3% of MO was degraded within 180 min, leading to complete degradation after 360 min in the 10th cycle. Based on the heterogeneity of the catalytic system, the TC catalysts can be easily separated from the reaction mixture by careful decantation of the aqueous solution or, more accurately, via centrifugation, followed by washing with water and ethanol. Thus, the stability of the TC50 catalyst was examined by performing ten consecutive catalytic tests by irradiating 5 ppm of MO in the presence of 0.2% w/v photocatalyst for 180 min under simulated solar light. As it can be seen in Figure 10, even after 10 cycles, 89.3% of MO was degraded within 180 min, leading to complete degradation after 360 min in the 10th cycle. The UV-Vis spectra and macroscopic evaluation before and after ten consecutive cycles of MO degradation using TC50 under visible light are illustrated in Figures S11 and S12. To the best of our knowledge, this result is far superior compared to already published works on TiO 2 /C-dots [23,25], showing a robust, almost eternal photocatalytic system for continuous pollution protection. The UV-Vis spectra and macroscopic evaluation before and after ten consecutive cycles of MO degradation using TC50 under visible light are illustrated in Figures S11 and S12. To the best of our knowledge, this result is far superior compared to already published works on TiO2/C-dots [23,25], showing a robust, almost eternal photocatalytic system for continuous pollution protection. Since the photodegradation of MO in the presence of TiO2/Cdots follows a pseudofirst-order kinetic model, photocatalytic efficiency of all TC25-75 nanocomposites, compared to TC0 and TAu, under different radiation sources can be summarized in Table 1. In all cases, the rate constants were obtained by plotting ln(C0/C) to irradiation time. As it can be seen, the degradation rate of MO for TC25 and TC50 under visible light irradiation is 12.5 times higher than that of pure TiO2 (TC0), while under UV-A and solar light irradiation, a 3.4-4.6 and 4.4-5.7 times higher rate was observed. Moreover, upon solar light irradiation, TC25 and TC50 provided a 3.3-4.3 higher degradation rate compared to commercially available Au/TiO2 (TAu) catalyst.  Since the photodegradation of MO in the presence of TiO 2 /Cdots follows a pseudofirst-order kinetic model, photocatalytic efficiency of all TC25-75 nanocomposites, compared to TC0 and TAu, under different radiation sources can be summarized in Table 1. In all cases, the rate constants were obtained by plotting ln(C 0 /C) to irradiation time. As it can be seen, the degradation rate of MO for TC25 and TC50 under visible light irradiation is 12.5 times higher than that of pure TiO 2 (TC0), while under UV-A and solar light irradiation, a 3.4-4.6 and 4.4-5.7 times higher rate was observed. Moreover, upon solar light irradiation, TC25 and TC50 provided a 3.3-4.3 higher degradation rate compared to commercially available Au/TiO 2 (TAu) catalyst.

Photocatalytic Activity of Multifunctional Protective Films
The photocatalytic ability of the catalysts was also tested after their incorporation into the silica-based consolidant FX-C and their application onto cementitious mortars. Mixtures containing 4% w/w of TC0, TC25 and TAu into the FX-C were prepared ( Figure S1) and brushed until saturation onto the surface of the cementitious specimens, which were previously cleaned with isopropanol. TC25 was selected on the grounds of its bests photocatalytic activity both under solar and UV irradiation, as it was proved in the experiments with the MO degradation. For comparison, a part of the surface of the samples was left untreated, and another one was brushed with the consolidant without a catalyst. In order to explore the most favorable application method, another testing surface was created by first applying a layer of the FX-C until saturation and, afterward, a layer of the mixture of the consolidant with TC25. In order to study the self-cleaning properties of the specimens with the different protective films, two drops (around 0.025 mL each) of methylene blue (MB) were released on each coating, and its degradation was evaluated by comparing the color difference of the stained surface with the color of the surface before the MB application ( Figure S2), using a spectrophotometer. The goal is to observe a decline in this color difference (∆E), reaching, in the best case, a value below 5. The macroscopic evaluation of MB degradation on treated and untreated cement mortars is reported in Figure S2. The specimens were tested under UV-A and visible light irradiation for 48 h, and the results do not significantly differ regarding the type of radiation used. As shown in Figure 11 and Table 2, the protective films containing TC0 and TAu were those with the best performance, reaching the desired value within 48 h.
In addition to the above-mentioned photocatalysts, the surface treated with a double layer consisting of FX-C and FX-C/TC25 also demonstrated considerable self-cleaning properties both under UV-A and visible light irradiation, reaching a value of 7.21 and 5.76, respectively. Moreover, in the case of the protective films containing TC25, the degradation of the pigment was also notable after 48 h of irradiation. The untreated surfaces and the surface treated with solely FX-C showed an insignificant decline in the color difference that can be attributed to the degradation of the dye because of the action of the irradiation. It should be mentioned that, despite the fact that the FX-C/TAu film demonstrated substantial photocatalytic properties, it cannot be proposed as a viable protective option for monumental surfaces, as it changes the color of the surface radically after its application ( Figure S2). each coating, and its degradation was evaluated by comparing the color difference of the stained surface with the color of the surface before the MB application ( Figure S2), using a spectrophotometer. The goal is to observe a decline in this color difference (ΔΕ), reaching, in the best case, a value below 5. The macroscopic evaluation of MB degradation on treated and untreated cement mortars is reported in Figure S2. The specimens were tested under UV-A and visible light irradiation for 48 h, and the results do not significantly differ regarding the type of radiation used. As shown in Figure 11 and Table 2, the protective  films containing TC0 and TAu were those with the best performance, reaching the desired  value within 48   These in situ experiments clearly pointed out that significant differences arose in the efficiency of the pure photocatalysts and the subsequent incorporation within protective coatings [65,66]. Regardless of the best performed TC25 and TC50 photocatalysts compared to TC0 and TAu under UV-A and solar irradiation, however, the nanocomposites that faster degraded the MB resulted in being those without C-dot loading into TiO 2 . Therefore, selfcleaning coatings applied to building materials require individual experiments, including a treatment process with both protective and self-cleaning agents. It was demonstrated that for cementitious surfaces, FX-C with pure TiO 2 and a double layer treatment of FX-C along with FX-C + TC25 could be considered promising solutions against pollutant decay. This study highlighted the need for in situ testing of the nanostructured protective coatings, in which the well-established photocatalysts were integrated for multifunctional building protection.

Conclusions
TiO 2 /C-dots composites with different C-dot loading were synthesized using citric acid, hydroxylamine and titanium isopropoxide (TTIP) by a facile, green and large-scale hydrothermal strategy. The different mass ratios of C-dots solution to TTIP produced TiO 2 functionalized photocatalysts that were characterized by the anatase structure, formed at a low calcination temperature of 200 • C. The C-dot modified TiO 2 showed efficient photocatalytic MO and MB degradation activity under UV-A and, more specifically, under visible light and sunlight radiation. Under UV-A, visible and solar light irradiation, the degradation rate of MO for TC25 and TC50 is 3.4-12.5 times higher compared to pure TiO 2 (TC0), while solar light provides a 3.3-4.3 higher degradation rate of MO for TC25 and TC50 than commercially available Au/TiO 2 (TAu) catalyst. Additionally, the photocatalysts can be recycled at least 10 times without losing their activity, showing the potentiality of these nanocomposites for continuous pollution protection. Finally, it was demonstrated that the application of multifunctional coatings on the cementitious matrix played a significant role. In the in situ applications of the self-cleaning coatings onto cementitious materials, the protective film of the silica-based FX-C with the pure TiO 2 and the double layer treatment with FX-C and FX-C + TC25 resulted in being the best-performing coatings against pollutant degradation.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/coatings12050587/s1, Table S1: Mix design of cement mortars treated with the nanocomposites of FX-C with catalysts; Figure S1: Sols of FX-C and nanocomposites of FX-C with catalysts; Figure S2: Macroscopical evaluation of the gradual MB degradation on treated and untreated cement mortars, before and after MB application, under (a) UV-A and (b) visible irradiation; Figure S3: (a,b) TEM images and (c) diameter distribution histogram of C-dots; Figure S4: (a) TEM image and (b) diameter distribution histogram of TC75; Figure S5: UV-Vis spectra of gradual MO degradation using TC50 upon UV-A irradiation; Figure S6: Macroscopic image of gradual MO degradation using TC50 upon UV-A irradiation; Figure S7: UV-Vis spectra of gradual MO degradation using TC50 upon simulated solar light irradiation; Figure S8: Macroscopic image of gradual MO degradation using TC50 upon simulated solar light irradiation; Figure S9: UV-Vis spectra of gradual MO degradation using TC50 under sunlight irradiation; Figure S10: Macroscopic image of gradual MO degradation using TC50 under sunlight irradiation; Figure S11: UV-Vis spectra before (0 min) and after ten consecutive cycles (1-10) of MO degradation using TC50 upon simulated solar light irradiation; Figure S12: Macroscopic image before (0 min) and after ten consecutive cycles (1-10) of MO degradation using TC50 upon simulated solar light irradiation.