Enhancing the Visible Light Photocatalytic Activity of TiO 2 -Based Coatings by the Addition of Exfoliated g-C 3 N 4

: In the last few years, increasing interest from researchers and companies has been shown in the development of photocatalytic coatings for air purification and self-cleaning applications. In order to maintain the photocatalyst’s concentration as low as possible, highly active materials and/or combinations of them are required. In this work, novel photocatalytic formulations containing g-C 3 N 4 /TiO 2 composites were prepared and deposited in the form of coatings on a-block substrates. The obtained photocatalytic surfaces were tested for NOx and acetaldehyde removal from model air. It was found that the addition of only 0.5 wt% g-C 3 N 4 towards TiO 2 content results in over 50% increase in the photocatalytic activity under visible light irradiation in comparison to pure TiO 2 coating, while the activity under UV light was not affected. The result was related to the creation of a g-C 3 N 4 /TiO 2 heterojunction that improves the light absorption and the separation of photogenerated electron-hole pairs, as well as to the inhibition of TiO 2 particles’ agglomeration due to the presence of g-C 3 N 4 sheets.


Introduction
The effect of outdoor and indoor pollutants on human health and the environment has been a matter of increasing importance during the last decades and therefore stricter legislations regarding emissions have been continuously applied.Among the most common pollutants, nitrogen oxides (NOx) [1,2], volatile organic compounds (VOCs) [3,4], and microorganisms [5,6] have been confirmed as significantly dangerous.Continuous exposure to them can cause serious respiratory problems, while at high concentrations they are considered particularly toxic.Construction materials in buildings are directly and constantly exposed to these kinds of pollutants [7], especially in enclosed spaces (such as workplaces, tunnels, underground parking, etc.) where a great accumulation of pollutants can occur.Importantly, these spaces' surfaces can serve as large workable sites where the pollutants can be decomposed on.Thus, the research community is focusing on developing techniques and materials to utilize these surfaces for efficient air purification.In this direction, semiconductor photocatalysis is increasingly attracting the attention of researchers and companies since it is regarded as a green and viable technology that is efficient under normal ambient conditions [8].After a photocatalyst is irradiated with UV or visible light, the generated electron-hole (e − -h + ) pairs are separated and can react with adsorbed water and/or oxygen from the atmosphere to form reactive oxygen species (ROS).Then, the ROS may react with the organic or inorganic molecules present on the surface of the photocatalyst, thus oxidizing and/or reducing them into harmless compounds.However, the photocatalysts cannot be used in their common powder form in real-life scenarios, as without proper immobilization they would drift away from the building surface.Therefore, photocatalytic building materials are fabricated by embedding the photocatalyst in the bulk of cements, paints, etc., or by applying the photocatalyst in the form of coatings on various surfaces [9] with the latter gaining increasing attention.
Considering that the photocatalytic reactions occur on the material surface and require direct contact between the photocatalyst and the pollutants, methods such as spray coating [10,11] and dip coating [12] need to be used to uniformly apply the photocatalytic material onto the substrate surface.In addition, to maintain the transmittance or the original color of the substrate, the coating layer should be highly transparent [13].This means that the coating has to be thin and the concentration of the photocatalyst as low as possible.Furthermore, the coating deposition on existing building surfaces should not be followed by any severe post-deposition treatment such as high-temperature calcination, irradiation, etc. Ideally, the coating should be deposited by a one-step procedure at normal conditions (temperature and pressure) to have long-lasting bonding with the building surface and facile maintenance [14].
Titanium dioxide (TiO 2 )-based coatings have been studied and used for air purification and self-cleaning applications [14][15][16][17].Undoubtedly, TiO 2 is currently the most used photocatalyst in photocatalytic construction materials due to its high chemical stability and relatively low price [18,19].Nonetheless, TiO 2 effectiveness is limited by its UV responsiveness owing to its wide band gap (E g = 3.2 eV) and fast e − -h + recombination caused by the high density of trap states [20,21].Specifically, for the air depollution from NOx (NO and NO 2 ) gases, novel photocatalysts are searched for in order to tackle the main disadvantages of the TiO 2 , i.e., UV sensitization and substantial release of toxic NO 2 product.Other metal oxides [22,23], inorganic substances like Ag 3 PO 4 [24], and organic polymers like g-C 3 N 4 [25,26] alone or in the form of composites have been reported as very efficient photocatalytic materials owing to their visible light sensitization and/or low emission of NO 2 .
Amongst the newly developed materials, graphitic carbon nitride (g-C 3 N 4 ) has emerged as a non-toxic, visible-light-driven (E g = 2.7 eV) photocatalyst with exceptional optical properties and stability [27].Over the past decade, g-C 3 N 4 has been studied for renewable energy utilization [28] and environmental depollution.g-C 3 N 4 has been proven promising for use in construction materials and coatings [29,30]; however, the expected wide practical application of the g-C 3 N 4 photocatalyst has not been reached yet, neither as powder nor in the form of a coating.
While g-C 3 N 4 /TiO 2 photocatalytic powders have been extensively investigated, g-C 3 N 4 /TiO 2 in the form of a coating is still poorly explored.Some reports on solar fuel photocatalytic production [31], water purification [32], and air treatment [33] can be mentioned here.Generally, the increase in the photocatalyst's amount improves the photocatalytic performance of coatings, however, their transmittance and homogeneity decrease [34]; thus, the great challenge remains on how to resolve this problem without compromising the optical and photocatalytic properties.
In this work, novel photocatalytic coatings containing g-C 3 N 4 /TiO 2 composites were prepared by adding small quantities of chemically exfoliated g-C 3 N 4 to a commercially available TiO 2 suspension of NanoPhos S.A. [35].The modified suspensions were then sprayed on a-block substrates and the obtained coatings were examined for outdoor and indoor air purification, specifically for photocatalytic removal of NOx and acetaldehyde pollutants under UV and visible light irradiation.The photocatalytic stability of the coatings was tested in order to evaluate their suitability for real-life applications; also, the physicochemical properties of the photocatalytic component, i.e., g-C 3 N 4 /TiO 2 composites in the form of powder, were examined.

Crystalline Structure and Chemical Composition
The XRD patterns of the obtained powder samples are presented in Figure 1.The pattern of chemically exfoliated g-C 3 N 4 is also given for comparison.All powder samples showed the characteristic for TiO 2 peaks attributed to its anatase (ICDD, PDF-2 #21-1272) and rutile (ICDD, PDF-2 #21-1276) phases [36].The patterns of the composite materials showed no distinct diffraction peaks related to the exfoliated g-C 3 N 4 .This can be ascribed to (i) its low content, i.e., 0.1, 0.5, 1.0, 2.0, and 5 wt% towards the TiO 2 ; (ii) its low crystallinity due to the performed chemical exfoliation; and (iii) to the fact that the main (002) diffraction peak of g-C 3 N 4 at 27.8 • -corresponding to the interlayer stacking of aromatic rings with a distance of d = 0.32 nm [27]-overlaps with the (110) diffraction peak of the rutile phase at 27.4 • [37].

Crystalline Structure and Chemical Composition
The XRD patterns of the obtained powder samples are presented in Figure 1.The pattern of chemically exfoliated g-C3N4 is also given for comparison.All powder samples showed the characteristic for TiO2 peaks attributed to its anatase (ICDD, PDF-2 #21-1272) and rutile (ICDD, PDF-2 #21-1276) phases [36].The patterns of the composite materials showed no distinct diffraction peaks related to the exfoliated g-C3N4.This can be ascribed to (i) its low content, i.e., 0.1, 0.5, 1.0, 2.0, and 5 wt% towards the TiO2; (ii) its low crystallinity due to the performed chemical exfoliation; and (iii) to the fact that the main (002) diffraction peak of g-C3N4 at 27.8°-corresponding to the interlayer stacking of aromatic rings with a distance of d = 0.32 nm [27]-overlaps with the (110) diffraction peak of the rutile phase at 27.4° [37].The presence of the g-C3N4 in the composites was revealed by the FT-IR spectra shown in Figure 2. Specifically, the peaks between 1230 and 1630 cm −1 are associated with the characteristic stretching of C-N heterocycles including trigonal N-(C)3 and bridging H-N-(C)2 units [38,39].Notably, the peaks become more prominent with the increase in the g-C3N4 concentration in the composites.Also, the broad peak from 400 to 700 cm −1 is attributed to the Ti-O stretching and Ti-O-Ti bridging stretching modes of TiO2 [40], thus confirming the presence of both components of the g-C3N4/TiO2 heterostructures.The observed broad peaks in the region from 3000 to 3500 cm −1 are attributed to N-H and O-H bonds originating from residual amino groups of the melamine precursor [41] and surface adsorbed water [42], respectively.The presence of the g-C 3 N 4 in the composites was revealed by the FT-IR spectra shown in Figure 2. Specifically, the peaks between 1230 and 1630 cm −1 are associated with the characteristic stretching of C-N heterocycles including trigonal N-(C) 3 and bridging H-N-(C) 2 units [38,39].Notably, the peaks become more prominent with the increase in the g-C 3 N 4 concentration in the composites.Also, the broad peak from 400 to 700 cm −1 is attributed to the Ti-O stretching and Ti-O-Ti bridging stretching modes of TiO 2 [40], thus confirming the presence of both components of the g-C 3 N 4 /TiO 2 heterostructures.The observed broad peaks in the region from 3000 to 3500 cm −1 are attributed to N-H and O-H bonds originating from residual amino groups of the melamine precursor [41] and surface adsorbed water [42], respectively.

Surface Area and Morphology
The liquid N2 adsorption-desorption isotherms and the pore size distribution of the powder samples are shown in Figure 3a,b, correspondingly.All materials demonstrated characteristic hysteresis loops of type IV isotherms typical for mesoporous materials [43].Regarding the pure components, the sample p-TiO2 showed a specific surface area (SSA) of 48.2 m 2 /g, while the exfoliated g-C3N4 exhibited a significantly higher SSA of 99.3 m 2 /g.The g-C3N4/TiO2 heterostructures exhibited SSA close to this of p-TiO2, which slightly increases with the increase in the g-C3N4 content.Namely, samples p-TiCN0.1,p-TiCN0.5, p-TiCN1, p-TiCN2, and p-TiCN5 showed SSA of 48.4,48.6, 48.9, 49.1, and 49.2 m 2 /g, respectively.The materials demonstrated similar pore size distribution with a broad peak at ~180 Å (radius 18 nm) typical for the commercial P25 TiO2 [44,45].The peak is indicative of small and large mesopores that correspond to pores within individual nanoparticles and spaces between the formed agglomerates, respectively [46].The exfoliated g-C3N4 exhibited wider pore size distribution, which can be related to its layered structure.
In Figure 4, two typical SEM images of the samples p-TiO2 and p-TiCN5 are presented.For sample p-TiO2, the characteristic morphology of TiO2 particles forming large agglomerates can be observed.In sample p-TiCN5, the lamellar sheets of exfoliated g-C3N4 can also be seen intermingled with the TiO2 agglomerates.In this case, the TiO2 agglomerates appear to be smaller in size.This is a key finding since the g-C3N4 sheets seem to prevent the excessive agglomeration of TiO2 particles.This is expected to play an important role in the overall photocatalytic activity by improving the homogeneity of the prepared formulation suspension and the deposited on the a-block photocatalytic coating.

Surface Area and Morphology
The liquid N 2 adsorption-desorption isotherms and the pore size distribution of the powder samples are shown in Figure 3a,b, correspondingly.All materials demonstrated characteristic hysteresis loops of type IV isotherms typical for mesoporous materials [43].Regarding the pure components, the sample p-TiO 2 showed a specific surface area (SSA) of 48.2 m 2 /g, while the exfoliated g-C 3 N 4 exhibited a significantly higher SSA of 99.3 m 2 /g.The g-C 3 N 4 /TiO 2 heterostructures exhibited SSA close to this of p-TiO 2 , which slightly increases with the increase in the g-C 3 N 4 content.Namely, samples p-TiCN0.1,p-TiCN0.5, p-TiCN1, p-TiCN2, and p-TiCN5 showed SSA of 48.4,48.6, 48.9, 49.1, and 49.2 m 2 /g, respectively.The materials demonstrated similar pore size distribution with a broad peak at ~180 Å (radius 18 nm) typical for the commercial P25 TiO 2 [44,45].The peak is indicative of small and large mesopores that correspond to pores within individual nanoparticles and spaces between the formed agglomerates, respectively [46].The exfoliated g-C 3 N 4 exhibited wider pore size distribution, which can be related to its layered structure.
In Figure 4, two typical SEM images of the samples p-TiO 2 and p-TiCN5 are presented.For sample p-TiO 2 , the characteristic morphology of TiO 2 particles forming large agglomerates can be observed.In sample p-TiCN5, the lamellar sheets of exfoliated g-C 3 N 4 can also be seen intermingled with the TiO 2 agglomerates.In this case, the TiO 2 agglomerates appear to be smaller in size.This is a key finding since the g-C 3 N 4 sheets seem to prevent the excessive agglomeration of TiO 2 particles.This is expected to play an important role in the overall photocatalytic activity by improving the homogeneity of the prepared formulation suspension and the deposited on the a-block photocatalytic coating.

Light Absorbance and Band Gap Estimation
The measured diffuse reflectance spectra (R) of the powder samples, as well as the spectrum of g-C3N4, are shown in Figure 5a.It is evident that the g-C3N4 exhibited absorbance in the visible region (between 390 and 420 nm), while the p-TiO2 and all the composite g-C3N4/TiO2 samples have similar spectra, with the main absorbance

Light Absorbance and Band Gap Estimation
The measured diffuse reflectance spectra (R) of the powder samples, as well as the spectrum of g-C 3 N 4 , are shown in Figure 5a.It is evident that the g-C 3 N 4 exhibited absorbance in the visible region (between 390 and 420 nm), while the p-TiO 2 and all the composite g-C 3 N 4 /TiO 2 samples have similar spectra, with the main absorbance positioned between 360 and 400 nm.By utilizing the Kubelka-Munk functions F = (1 − R) 2 /2R, the corresponding absorption functions (F × E) n = f(E) were constructed, as depicted in Figure 5b.The two semiconductors are considered to possess an indirect band gap, and to that end, n = 1/2 was used.The band gap energy (E g ) of the materials was estimated by extrapolation of the linear part of the absorption functions to the x-axis that is the photon energy in eV [47].It can be seen that the E g of the g-C 3 N 4 is 2.98 eV due to its exfoliated nature, while the E g of the p-TiO 2 (3.05 eV) and all the composite g-C 3 N 4 /TiO 2 samples are similar (3.02-3.03eV).The results prove that the addition of g-C 3 N 4 in low concentrations does not significantly affect the E g of the materials.

Light Absorbance and Band Gap Estimation
The measured diffuse reflectance spectra (R) of the powder samples, as well as the spectrum of g-C3N4, are shown in Figure 5a.It is evident that the g-C3N4 exhibited absorbance in the visible region (between 390 and 420 nm), while the p-TiO2 and all the composite g-C3N4/TiO2 samples have similar spectra, with the main absorbance positioned between 360 and 400 nm.By utilizing the Kubelka-Munk functions F = (1 − R) 2 /2R, the corresponding absorption functions (F × E) n = f(E) were constructed, as depicted in Figure 5b.The two semiconductors are considered to possess an indirect band gap, and to that end, n = 1/2 was used.The band gap energy (Eg) of the materials was estimated by extrapolation of the linear part of the absorption functions to the x-axis that is the photon energy in eV [47].It can be seen that the Eg of the g-C3N4 is 2.98 eV due to its exfoliated nature, while the Eg of the p-TiO2 (3.05 eV) and all the composite g-C3N4/TiO2 samples are similar (3.02-3.03eV).The results prove that the addition of g-C3N4 in low concentrations does not significantly affect the Eg of the materials.

Photocatalytic Activity
The variation in the gases' (NO, NO2, and NOx) concentrations over the deposited photocatalytic coatings under dark and during light irradiation can be seen in the recorded experimental concentration curves in Figure S1.In Figure 6, the calculated changes in the gas concentrations, i.e., the decrease in the NO concentration, the increase

Photocatalytic Activity
The variation in the gases' (NO, NO 2 , and NOx) concentrations over the deposited photocatalytic coatings under dark and during light irradiation can be seen in the recorded experimental concentration curves in Figure S1.In Figure 6, the calculated changes in the gas concentrations, i.e., the decrease in the NO concentration, the increase in the NO 2 concentration, and the total decrease in the NOx concentration in the air with regard to their initial concentrations (C NO = 1 ppm, C NO2 = 0 ppm, and C NOx = 1 ppm) are presented.Under UV light irradiation (Figure 6a), all coatings showed comparable photocatalytic activity with a slightly better performance of the coating with 0.5 wt% g-C 3 N 4 (sample c-TiCN0.5)achieving 48% NOx removal.However, under visible light irradiation (Figure 6b), the addition of g-C 3 N 4 led to significantly enhanced activity of the coatings.The coating with 1 wt% g-C 3 N 4 (sample c-TiCN1) reached 12% total NOx removal, which is 4.5% higher than the coating with pure TiO 2 (sample c-TiO 2 ).This increase designates the important contribution of g-C 3 N 4 in broadening the light response from UV to the visible light region [37], which is essential for the real-life application of the photocatalyst.The results for the photocatalytic acetaldehyde oxidation by the coatings under UV and visible light irradiation are presented in Figure 7, while the recorded experimental concentration curves can be seen in Figures S2 and S3.It can be observed that the sample c-TiCN0.5 exhibited the best photocatalytic activity under both UV and visible light The results for the photocatalytic acetaldehyde oxidation by the coatings under UV and visible light irradiation are presented in Figure 7, while the recorded experimental concentration curves can be seen in Figures S2 and S3.It can be observed that the sample c-TiCN0.5 exhibited the best photocatalytic activity under both UV and visible light irradiation, reaching 57% and 7.5% acetaldehyde removal, respectively.Again, the difference was more prominent under visible light where the addition of 0.5 wt% g-C 3 N 4 (sample c-TiCN0.5)led to ~4% higher removal than the coating with pure c-TiO 2 (sample c-TiO 2 ).In the present study, it is shown that the addition of small quantities of g-C3N4 to TiO2-based formulations improves significantly the activity under visible light irradiation.Notably, major physicochemical properties influencing the photocatalytic efficiency, i.e., SSA and porosity, as well as light absorbance and width of the band gap were only slightly affected by the addition of g-C3N4.We would like to emphasize that the increased photocatalytic activity can be attributed to two factors: (i) the creation of a g-C3N4/TiO2 heterojunction that enhances the charge (e − and h + ) separation; and (ii) the reduced agglomeration of the TiO2 nanoparticles.
More specifically, the coupling between g-C3N4 and TiO2 (P25) investigated in our previous work [48] revealed that the positions of the band gap's edges of the two In the present study, it is shown that the addition of small quantities of g-C 3 N 4 to TiO 2 -based formulations improves significantly the activity under visible light irradiation.Notably, major physicochemical properties influencing the photocatalytic efficiency, i.e., SSA and porosity, as well as light absorbance and width of the band gap were only slightly affected by the addition of g-C 3 N 4 .We would like to emphasize that the increased photocatalytic activity can be attributed to two factors: (i) the creation of a g-C 3 N 4 /TiO 2 heterojunction that enhances the charge (e − and h + ) separation; and (ii) the reduced agglomeration of the TiO 2 nanoparticles.
More specifically, the coupling between g-C 3 N 4 and TiO 2 (P25) investigated in our previous work [48] revealed that the positions of the band gap's edges of the two semiconductors in the energy scale favor the formation of a heterojunction with band alignment type II.Then, taking into account the potentials for •OH and •O 2 radicals' formation, the most likely charge transfer mechanism is the Z-scheme [49,50], which facilitates the charge separation and reduces the recombination rate of photogenerated e − -h + pairs [51][52][53].It should be noted here that, as it comprises two TiO 2 phases (anatase and rutile), the P25 photocatalyst possesses its own charge separation mechanism [54,55], which results in enhanced formation of •OH radicals on the nanoparticles' surface.Through the Z-scheme, the g-C 3 N 4 component provides an additional pathway for the e − generated in the conduction band of the TiO 2 component (CB TiO2 ) towards its valence band (VB g-C3N4 ) and finally to its conduction band (CB g-C3N4 ).Thus, the charge separation in the g-C 3 N 4 /TiO 2 heterojunction and the formation of •OH radicals on the composites' surface are further enhanced.In relation to this, our previous EPR results on the formation of reactive oxygen species in g-C 3 N 4 /TiO 2 systems with various ratios [56] also demonstrated that •OH radicals are predominately formed in TiO 2 -rich systems under visible light irradiation.Considering also that NOx and acetaldehyde oxidation mechanisms are highly dependent on •OH radicals, the enhanced photocatalytic removal of NOx and acetaldehyde recorded in this study can be attributed to the favorable charge transfer pathways and separation in the g-C 3 N 4 /TO 2 heterostructures with low g-C 3 N 4 content in comparison with pure p-TiO 2 .
The reduced agglomeration of the TiO 2 nanoparticles-and therefore the improved homogeneity of the photocatalytic material in the formulation suspension-leads to the uniform distribution and better exposure of photocatalyst on the coated surfaces, facilitating the interaction of the gas pollutants with more photocatalytic centers.However, when the content of g-C 3 N 4 exceeds a certain value (over 1 wt% towards TiO 2 ), the activity of the photocatalytic coatings decreases.This occurs because TiO 2 becomes covered by the g-C 3 N 4 sheets, thus reducing the irradiation amount reaching the semiconductor.Also, according to [57,58], the photogenerated electron-hole pairs cannot reach the surface of the g-C 3 N 4 /TiO 2 composite as the g-C 3 N 4 layers become recombination centers.g-C 3 N 4 /TiO 2 with 0.5-1.0wt% g-C 3 N 4 towards TiO 2 appeared as the optimum ratios for a suitable balance of these processes, leading to higher photocatalytic activity of the composites in comparison to the pure TiO 2 coating.
The finding that addition of only 0.5 wt% g-C 3 N 4 towards TiO 2 to a well-established commercial formulation of NanoPhos S.A. results in over 50% increase in the photocatalytic activity under visible light irradiation makes the coatings suitable for immediate practical application, especially for uses where the UV light is negligible (i.e., indoor applications), while at the same time, the production cost is kept as low as possible.
Being considered for real-life application, the c-TiCN0.5 sample was subjected to weathering in order to examine its long-term stability.The results shown in Figures 8 and 9 demonstrate that after each 500-h cycle, the coating exhibited a decrease in photocatalytic activity, both for NOx and acetaldehyde removal, which is attributed to the wear and damage of the thin photocatalytic coating due to the intense weathering conditions.Critically though, even after three cycles of weathering, the coating remains active.The photocatalytic and weathering stability suggest a good adhesion of the coating to the a-block surface, which is extremely important for commercial application as no additional post-deposition treatment (usually necessary for the stabilization of coatings on a surface) is required.Considering the real-life application potential, it is established that the coating product maintains its value for approximately 18 to 36 months, which is well within the range of other commercial products.

Preparation of Materials and Coatings
Melamine powder was purchased from Alfa Aesar.Sulfuric acid (H2SO4, 98%) was purchased from Chem-Lab and isopropanol (IPA) from Sigma-Aldrich (Saint Louis, MO, USA).The chemicals were of analytical grade and were used without any further purification.The syntheses of bulk and chemically exfoliated g-C3N4 are described in detail in our previous work [59].In brief, 4 g of melamine was put into an open alumina crucible and treated at 550 °C for 3 h in a tube furnace under Argon flow.The resulting yellow bulk g-C3N4 was ground into powder and subjected to chemical exfoliation.Namely, 0.5 g of bulk g-C3N4 was stirred in 8 mL concentrated H2SO4 for 3 h.Then, 2 mL distilled H2O was slowly added and the suspension was poured into 100 mL distilled H2O.After centrifugation and washing until pH 7, the remaining white powder was re-

Preparation of Materials and Coatings
Melamine powder was purchased from Alfa Aesar.Sulfuric acid (H 2 SO 4 , 98%) was purchased from Chem-Lab and isopropanol (IPA) from Sigma-Aldrich (Saint Louis, MO, USA).The chemicals were of analytical grade and were used without any further purification.The syntheses of bulk and chemically exfoliated g-C 3 N 4 are described in detail in our previous work [59].In brief, 4 g of melamine was put into an open alumina crucible and treated at 550 • C for 3 h in a tube furnace under Argon flow.The resulting yellow bulk g-C 3 N 4 was ground into powder and subjected to chemical exfoliation.Namely, 0.5 g of bulk g-C 3 N 4 was stirred in 8 mL concentrated H 2 SO 4 for 3 h.Then, 2 mL distilled H 2 O was slowly added and the suspension was poured into 100 mL distilled H 2 O.After centrifugation and washing until pH 7, the remaining white powder was re-dispersed into 100 mL IPA and ultrasonicated for 1 h.Finally, the suspension was dried at 60 • C, thus obtaining exfoliated g-C 3 N 4 powder.
The photocatalytic dispersions were prepared by adding a specific amount of exfoliated g-C 3 N 4 in a commercially available TiO 2 -containing (2 wt%) formulation provided from NanoPhos S.A.The product is patented, while coatings deposited on a glass surface were examined in our previous work [16].In this study, the amounts of the g-C 3 N 4 were calculated, taking into account the known amount of TiO 2 in the formulation suspension (Table 1).Quantities corresponding to 0.1, 0.5, 1, 2, and 5 wt% of g-C 3 N 4 towards TiO 2 were added and the modified suspensions were sonicated for 15 min using a 1000 W probe-type Hielscher UIP1000hdT ultrasonic device operating at 20 kHz (Wanaque, NJ, USA).Then, the suspensions were deposited on 5 cm × 10 cm × 0.5 cm a-block substrates (coated surface 50 cm 2 ) using a Z-020-Tornador BLACK sprayer, (International Tool Company, Leicester, UK).The a-block is a lightweight, precast, foam-like concrete construction material with a porous surface.Its use is considered very beneficial as it insulates better than conventional bricks, while at the same time, it does not strain the building, reduces its load, and can withstand heavy constructions such as roofs.It should be mentioned that the coating was transparent and thin enough so that the coated and non-coated a-block substrates were similar in color (Figure 10).The coated substrates were left for curing at ambient conditions for at least 24 h before the photocatalytic tests.The photocatalytic coatings were named c-TiCNx, where c stands for coating and x is the wt% of g-C 3 N 4 towards TiO 2 .provided from NanoPhos S.A.The product is patented, while coatings deposited on a glass surface were examined in our previous work [16].In this study, the amounts of the g-C3N4 were calculated, taking into account the known amount of TiO2 in the formulation suspension (Table 1).Quantities corresponding to 0.1, 0.5, 1, 2, and 5 wt% of g-C3N4 towards TiO2 were added and the modified suspensions were sonicated for 15 min using a 1000 W probe-type Hielscher UIP1000hdT ultrasonic device operating at 20 kHz (Wanaque, NJ, USA).Then, the suspensions were deposited on 5 cm × 10 cm × 0.5 cm ablock substrates (coated surface 50 cm 2 ) using a Z-020-Tornador BLACK sprayer, (International Tool Company, Leicester, UK).The a-block is a lightweight, precast, foamlike concrete construction material with a porous surface.Its use is considered very beneficial as it insulates better than conventional bricks, while at the same time, it does not strain the building, reduces its load, and can withstand heavy constructions such as roofs.It should be mentioned that the coating was transparent and thin enough so that the coated and non-coated a-block substrates were similar in color (Figure 10).The coated substrates were left for curing at ambient conditions for at least 24 h before the photocatalytic tests.The photocatalytic coatings were named c-TiCNx, where c stands for coating and x is the wt% of g-C3N4 towards TiO2.Due to the type of the substrate, physicochemical properties such as crystalline structure, specific surface area, light absorption, etc., were determined for the photocatalytic component alone.For this purpose, g-C3N4/TiO2 powders were obtained by drying 5 g of the suspensions in a ventilated oven at 105 °C for 48 h.Then, the solids were collected and ground into fine powders.The samples were named p-TiCNx, where p stands for powder and x is the wt% of g-C3N4 towards TiO2.Due to the type of the substrate, physicochemical properties such as crystalline structure, specific surface area, light absorption, etc., were determined for the photocatalytic component alone.For this purpose, g-C 3 N 4 /TiO 2 powders were obtained by drying 5 g of the suspensions in a ventilated oven at 105 • C for 48 h.Then, the solids were collected and ground into fine powders.The samples were named p-TiCNx, where p stands for powder and x is the wt% of g-C 3 N 4 towards TiO 2 .

Characterization
A Siemens (Plano, TX, USA) D500 X-ray diffractometer was used for the XRD measurements.The FT-IR spectra of the materials were measured on a Nicolet (Wausau, WI, USA) iS50 spectrometer.The liquid N 2 adsorption-desorption isotherms were obtained with a Quantachrome (Boynton Beach, FL, USA) Autosorb-iQ instrument.SEM characterization was performed using an FEI (Hillsboro, OR, USA) Quanta Inspect Microscope equipped with a tungsten filament operating at 25 kV.The UV-Vis diffuse reflectance spectra were recorded by a Shimadzu (Tokyo, Japan) UV-2100 spectrophotometer using BaSO 4 as reference.

Photocatalytic Activity Evaluation
The photocatalytic activity of the coatings was measured for NOx removal from model gas and acetaldehyde oxidation under UV and visible light irradiation at room temperature.For the experiments, Philips (Andover, MA, USA) Cleo Compact 15 W lamps with an intensity of 10 W/m 2 and Nordex (Chicago, IL, USA) T5-8W-4000 K lamps with an intensity of 7000 lux were used, respectively.

NOx Removal
The photocatalytic activity of the coatings for NOx removal was evaluated using a standard procedure based on ISO/DIS 22197-1 [60].The setup configuration and experimental conditions are described in detail in our previous works [56,59].Particularly, the prepared coated substrates were placed in a continuous flow reactor with dimensions 30 cm × 5 cm.The concentrations of NO, NO 2 , and NOx were monitored using a Horiba APNA-370 chemiluminescence-based NOx analyzer (with a lower detectable limit of 0.0005 ppm) equipped with an O 3 lamp and a silicon photodiode sensor.

Acetaldehyde Oxidation
The photocatalytic activity of the coatings for acetaldehyde oxidation was evaluated using a standard procedure based on ISO/DIS 22197-2 [61].Again, the setup configuration and experimental conditions are described in detail in our previous work [54].The coated substrates were placed in a continuous flow reactor with dimensions 30 cm × 5 cm, while the concentration of acetaldehyde gas was monitored using a Shimadzu Tracera GC-2010 gas chromatograph equipped with a BID-2010 detector.

Photocatalytic Stability
The photocatalytic stability of the coatings was evaluated by performing consecutive weathering cycles using the QUV Accelerated Weathering Tester instrument by Q-Lab, which in a few days can reproduce the damage that occurs over months of outdoor exposure.To simulate outdoor weathering, the materials are subjected to alternating treatment with UV light (UVA-340 lamps at 0.71 W/m 2 ), condensing humidity (at 50 • C), and water spray (with rate of 5 L/min), at controlled temperatures.It is estimated that a 500 h cycle corresponds to 6-12 months of real outdoor exposure depending on the geographic location and climate characteristics.In this work, 3 cycles of 500 h were performed and after each cycle the photocatalytic activity of the coating was measured.

Conclusions
Novel photocatalytic coatings were prepared by adding small quantities of exfoliated g-C 3 N 4 in a commercially available TiO 2 suspension and consequent deposition on construction material known as a-block.Under visible light irradiation, the coatings showed significantly increased photocatalytic activity in comparison to the commercial TiO 2 -containing coatings.The result was attributed to (i) the creation of a g-C 3 N 4 /TiO 2 heterojunction that enhances the charge separation and reduces the recombination of the photogenerated electron-hole pairs; and (ii) the incorporation of g-C 3 N 4 , which inhibits the TiO 2 particles' agglomeration, thus facilitating the interaction of the gas pollutants with the photocatalytic centers.It was found that only 0.5-1.0wt% of g-C 3 N 4 towards the TiO 2 content in the suspension is required, probably due to its exfoliated nature and high specific surface area.The fact that the color of the substrate is not affected and the production cost is kept low makes the coatings very attractive for practical implementation.Additionally, the coated a-block surfaces showed significant photocatalytic and weathering stability owed to the good adhesion of the coating to the substrate surface.
Overall, the developed photocatalytic coatings are very promising for real-life applications, as they display exceptional results with negligible extra cost and without the need for additional procedures.Hopefully, this work will inspire further research and contribute to the adoption of exfoliated g-C 3 N 4 for commercial and practical applications.

Figure 1 .
Figure 1.XRD patterns of the composite g-C3N4/TiO2 powder samples and the pure TiO2 and exfoliated g-C3N4.

Figure 1 .
Figure 1.XRD patterns of the composite g-C 3 N 4 /TiO 2 powder samples and the pure TiO 2 and exfoliated g-C 3 N 4 .

Figure 2 .
Figure 2. FT-IR spectra of the composite g-C3N4/TiO2 powder samples and the pure TiO2 and exfoliated g-C3N4.

Figure 2 .
Figure 2. FT-IR spectra of the composite g-C 3 N 4 /TiO 2 powder samples and the pure TiO 2 and exfoliated g-C 3 N 4 .

Catalysts 2024 , 17 Figure 6 .
Figure 6.Photocatalytic activity of the coatings for NOx removal under UV (a) and visible light (b) irradiation.

Figure 6 .
Figure 6.Photocatalytic activity of the coatings for NOx removal under UV (a) and visible light (b) irradiation.

Figure 7 .
Figure 7. Photocatalytic activity of the coatings for acetaldehyde oxidation under UV (a) and visible light (b) irradiation.

Figure 7 .
Figure 7. Photocatalytic activity of the coatings for acetaldehyde oxidation under UV (a) and visible light (b) irradiation.

Figure 9 .
Figure 9. Photocatalytic stability of the c-TiCN0.5 coating for acetaldehyde oxidation under UV (a) and visible light (b) irradiation, after three consecutive weathering cycles.

Figure 9 .
Figure 9. Photocatalytic stability of the c-TiCN0.5 coating for acetaldehyde oxidation under UV (a) and visible light (b) irradiation, after three consecutive weathering cycles.

Table 1 .
Experimental parameters for the preparation of the coatings: wt% of g-C 3 N 4 towards TiO 2 , wt% of g-C 3 N 4 in the suspension, and density (d) of the suspension.

Table 1 .
Experimental parameters for the preparation of the coatings: wt% of g-C3N4 towards TiO2, wt% of g-C3N4 in the suspension, and density (d) of the suspension.