SELF-CLEANING TITANIA COATINGS FOR BETTER PRESERVATION OF ARCHITECTURAL HERITAGE

SELF-CLEANING TITANIA COATINGS FOR BETTER PRESERVATION OF ARCHITECTURAL HERITAGE Enrico Quagliarini1, Federica Bondioli2, Placido Munafò1, Giovanni Battista Goffredo1, Luca Riderelli3 1 DICEA Department of Architecture, Building and Civil Engineering, Polytechnic University of Marche, via Brecce Bianche, 60131, Ancona, Italy, e.quagliarini@univpm.it 2 DIMA Department of Materials and Environmental Engineering, University of Modena and Reggio Emilia, via Vignolese 905, 41100, Modena, Italy, federica.bondioli@unimore.it 3 SIMAU Department of Materials, Environmental Sciences and Urban Planning, Polytechnic University of Marche, via Brecce Bianche, 60131, Ancona, Italy, l.riderelli@univpm.it ABSTRACT Self-cleaning treatments can be a promising tool in order to better preserve the original aspect and characteristics of stone surfaces and Architectural Heritage. In addition, the realization of easy-to-clean surfaces limits cleaning and maintenance actions, thus reducing their costs. Titanium dioxide nano-particles can be used to obtain transparent self-cleaning coatings degrading and removing stains and polluting agents thanks to its photoinduced properties: super-hydrophilicity and photocatalicity. Aqueous TiO2 solution was deposited in two different amounts by spray coating on travertine, a porous limestone largely used in historical and monumental building, obtaining a single-layer coating and a three-layers coating. The present work was carried out in order to verify the effectiveness of titania sols and the absence of adverse effects on treated historical stone surfaces by appearance, wettability and self-cleaning analyses. Obtained results were promising for future applications: the transparency of the coatings, the absence of significant changes in the morphology of limestone and the self-cleaning efficiency seem to allow the use of these coatings on historical and architectural surfaces made up by travertine. Further analyses are currently under way in order to better evaluate durability and feasibility of use of these coatings in the field of stone conservation. Parole chiave/Key-words: Self-cleaning surfaces, Architectural Heritage, titanium dioxide, stone surfaces, photocatalysis Introduction Surface treatments can improve conservation, protection and maintenance of several Cultural Heritage elements, preserving their original appearance and characteristics, since the action of many degradation agents begins from the outer layers of these surfaces. Preventive measures on historical stone surfaces of monuments and Architectural Heritage exposed to outdoor environment could lead to evident and important benefits in their preservation, especially in aggressive urban atmosphere. The use of self-cleaning treatments over stones could realize a preventive protection system, making the conservation of the original aspect of treated elements easier, decreasing the deposition of pollutants and soiling on historical surfaces and reducing the onset of external degradation processes due to air pollution, soluble salts, black crusts formation, soiling phenomena and biochemical deterioration. Photocatalysis, the catalysis of chemical redox reactions activated by solar light, is a promising tool to realize self-cleaning surfaces (Chen et al 2009). This photoinduced effect can be used to photo-decompose both organic and inorganic polluting substances absorbed or deposited on photocatalytic coatings under mere solar light exposure. Titanium dioxide (TiO2) is one of the most used and efficient photocatalytic material, because of its outstanding efficiency, inexpensiveness and compatibility with a large number of different materials. It has been used to develop innovative solutions in different fields: water and air purification, anti-bacterial and selfsterilizing surfaces, food industry, cosmetics and building materials. The selfcleaning effect of TiO2 is used in a lot of building elements, as cement mortars, exterior tiles, paving blocks, glasses, paints, finishing coatings, road-blocks, metal elements, concrete pavements (Chen et al 2009, Fujishima et al 1972, Diamanti et al 2008, Zhao et al 2008, Bondioli et al 2009). By the use of TiO2 nano-particles it is possible to realize transparent self-cleaning coatings (Potenza et al 2007, Luvidi et al 2010), obtaining an active and preventive protection system over architectural surfaces activated by mere solar light. This solution could be very important to limit the number and cost of cleaning and maintenance interventions. Titanium dioxide activates its photocatalytic property under ultraviolet (UV) component of solar light spectrum. UV exposure also activates TiO2 superhydrophilism (Fujishima et al 2000), thus creating an uniform water film on treated surfaces. The formation of a water layer can prevent contact between external dirt and surface itself and make removal of degradation agents easier. Anyway hydrophilic property of TiO2 could bring to longer contact with water and higher water absorption, a potential source of damage for porous stones, so its use must be carefully evaluated. The creation of a water film over treated surfaces and the degradation of contaminants can bring to a real self-cleaning effect. The aim of this work is to investigate the possible use of titanium dioxide coatings in the field of Architectural Heritage, analyzing both their efficiency and the effects induced on treated substrates by their properties. Materials and methods The TiO2 photoactivity is greatly influenced by its microstructure. Amorphous titania shows poor photocatalytic property, while crystal structures of titanium dioxide (anatase, rutile and brookite) have very higher photoactivity, especially in presence of anatase phase. It is possible synthesize titania sols using different methods, one of the most investigated is sol-gel method. An aqueous titania sol was prepared through sol-gel technique starting from tetrapropyl orthotitanate (TPOT) precursor added dropwise to a bihydrate oxalic acid water solution. After a stirring and heating process a solution with an amorphous TiO2 content of 1 wt% was obtained. The obtained titania sol was subsequently hydrothermally crystallized at 134 °C, 2 bar, for 30 minutes in order to produce TiO2 anatase nano-crystals directly in the liquid phase without any other subsequent thermal process after deposition on surfaces, since in most cases it can be harmful for stone substrate and not compatible with other conservation treatments (Liciulli et al 2011, Quagliarini et al 2011, Quagliarini et al 2012). Xray diffraction analysis (XRD) confirmed the formation of anatase crystals having a 4 nm average size. After a partial aggregation nano-particles' dimensions were approximately from 40 to 50 nm, as determined by Dynamic Light Scattering analysis (DLS). Travertine, a porous natural limestone, has been selected as the reference substrate. This carbonatic stone was widely used in monumental elements, sculptures and as an important building material especially in the Mediterrean Basin and it is still very common in modern architecture all over the world. Travertine specimens (dimensions: 8.0 x 8.0 x 1.5 cm3) were prepared, according to the requirements of the performing tests. Obtained solution was directly deposited on travertine surfaces by spray coating, applying two different amounts of titania on the surfaces, thus obtaining a single layer treatment (T1) with a deposited amount of titania of 0.12 g/m2, and a three layers (T2) treatment with 0.40 g/m2 of deposited TiO2. After a drying phase in a ventilated oven (1 hour, 70 °C), only the titania nano-particles were adherent to the limestone surfaces since water of aqueous solution evaporated. This heating phase is not strictly necessary and it simply accelerates the normal process of drying. The coatings were observed by scanning electron microscope (SEM) and analyzed by energy-dispersive X-ray spectroscopy (EDS) in order to define the microstructure of coatings deposited on travertine substrate and to confirm the presence of TiO2 on the stone surfaces. The use of self-cleaning titania coatings in the field of Architectural Heritage is subject to the fulfillment of various requirements: TiO2 treatments must show real self-cleaning efficiency without greatly alter the original aspect of treated stone surfaces (transparency) and without produce possible harmful side effects, especially higher water penetration due to photo-induced super-hydrophilicity. As for maintenance of the original aspect of treated surfaces, both colour and gloss analyses were carried out. Chromatic values were evaluated according to European standard rules, using CieL*a*b* colour space, by a Konica Minolta CM-2600d spectrophotometer. Following European test (UNI EN 15866:2010), for each treatment (T1 and T2) four samples were analyzed and the measurements were repeated at least seven times. The measurement points were localized by a reference spatial grid to ensure precise repeated measurements in the same points. The gloss was measured by the use of a Novo Gloss Trio apparatus (Rophoint Instruments) using 60° as standard geometry, carrying out at least four measurements for each sample. Since changes in wettability due to photo-induced hydrophilicity could bring to greater water penetration and more exposure to soluble salts, acidic/basic compounds and degradation agents, water absorption was investigated using different methods. Static contact angle (CA) values were analyzed through a OCA 20 apparatus (DataPhysic Instrument GmbH) following UNI EN 15802:2010 standard rule. The measurements were carried out on 3 uncoated samples and 2 specimens for each kind of different treatment (T1 and T2), performing 15 measurements for each test sample using 5 µl volume water drops. The determination of static contact angles for porous and rough surfaces, such as that of travertine, is not simple and the results should be carefully considered. The analysis was repeated under UV light (exposure time: 50 minutes, wavelength range: 325-390 nm, irradiance value: 20 W/m2) in order to evaluate the effects of photo-induced hydrophilicity of TiO2. To evaluate if the different surface wettability of treated surfaces changes the water absorption of the material, capillary water absorption test was performed (UNI EN 15801:2010). Original travertine samples (8.0 x 8.0 x 1.5 cm3) were divided by cut into 3 parts each (about 2.5 x 8.0 x 1.5 cm3) and then their lateral faces were sealed to reduce water absorption from untreated faces of travertine during test, obtaining six samples for each treatment condition (untreated, T1, T2). After a drying phase in a ventilated oven to reach constant mass m0 (difference between two successive weighing at an interval of 24 hours is not greater than 0.1 % of the mass of the specimen), stone specimens were put inside a vessel in contact with a bedding layer saturated with demineralized water. The amount of absorbed water at time ti (s) was measured by weighing and then denoted as absorbed mass per unit area Qi (kg/m2): Qi = (mi – m0)/A where A is the area of the stone samples exposed to water (m2). In-house test was prepared to evaluate surface water absorption. A settled amount of deionised water was sprayed every 2 minutes by the use of a manual nebulizer on the surface of the stone specimens placed on a support at 30 cm of distance and inclined 10 degrees from the vertical plane up to a total of about 45 ml. The analysis of surface absorption of water sprayed by manual nebulizer was carried out for one hour, measuring sample weight every 2 minutes. Same specimens used for capillary water absorption test were analyzed, 3 for each condition (untreated, T1, T2). The test was carried out with and without ultraviolet illumination (exposure time: 60 minutes, wavelength range: 325-390 nm, irradiance value: 20 W/m2) to assess the consequences of photo-induced hydrophilicity. In order to evaluate the self-cleaning ability of the coating, two different test were carried out, following Italian standard rules. Photocatalytic efficiency against pollutants was determined by degradation of nitrogen oxide during a continuous flow test method under UV light (UNI 11247:2010). Samples were placed in a 3 l borosilicate reactor in which dry air containing approximately 0.6 parts per million (ppm) of NO was passed through at a rate of 1.5 l/min. Subsequently surfaces were exposed to UV-A irradiation (irradiance value: 20 W/m2) for at least 45 minutes. Photocatalytic decomposition was monitored every minute up to a total of about 120 minutes, by a Nitrogen Oxides Analyzer 8841 (Rancon Instruments). A dye photo-decomposition test (UNI 11259:2008) was used to evaluate the degradation of soil and dirt under ultraviolet light by colorimetric method. Stone surfaces were stained by a rhodamine B water solution (rhodamine B content: 0.05 ± 0.005 g/l, 0.5 ml of solution per specimen, 2 samples for each type of condition) and then, after a 24 hours long drying phase, the samples were exposed to UV-A light (irradiance value: 3.75 ± 0.25 W/m2). The degradation of rhodamine B deposited on travertine surfaces was evaluated by colorimetry before UV irradiation and after 4 and 26 hours of exposure. According to the UNI standard rule, due to the red colour of rhodamine, only chromatic coordinate a* was used to assess the photocatalytic effect of self-cleaning coatings. Results and discussion Microstructure of titania coatings does not greatly alters the substrate morphology. The analysis showed that thin titanium dioxide film (far below 1 micron thickness) covers stone surfaces without originating cracks and segregation or modifying substrate surface in a large way (Fig. 1). The presence of TiO2 adherent to stone substrate was confirmed by EDS analysis and, as expected, the measured titania amount was higher in multilayer coating case (Fig. 1). Transparency of the coatings was evaluated by colour and gloss analyses. The results of colour changes ∆E* due to treatment deposition are shown in Figure 2: variations (∆E*T1 = 1.4, ∆E*T2 = 2.0) are negligible, undetectable by naked eye. The hue variations are satisfactory in the field of Architectural Heritage and they seem to be partly dependent on the TiO2 content, since colour variation ΔE* increases as titania content increases even if not in a proportional way. Gloss was not altered at all by TiO2 coatings: its value was 2.5 GU for all surfaces. The use of analyzed TiO2 solution seems to be compatible with historical stone surfaces, since coatings are transparent and the travertine samples maintain their original aspect. Several analyses evaluated the presence of well-known photo-induced hydrophilicity of titanium dioxide. Water static contact angle values without UV light exposure are not different between treated and untreated surfaces and measured variations, characterized by quite high standard deviation values, are mainly related to the physical heterogeneous properties (porosity and roughness) of limestone substrates. In comparison with untreated case, average contact value of T1 surfaces showed little increase while T2 treatment showed no changes at all. Hydrophilicity of titania is well evident under UV irradiation and contact angle values of treated surfaces become evidently lower as the time of UV light exposure increases (Fig. 3). Static contact angles of treated surfaces decrement especially during early minutes of UV irradiation and after 30 minutes of UV exposure the values are very similar in both cases (T1 and T2), decreasing drastically till the final almost stable results (Fig. 3). Treated surfaces clearly showed a hydrophilic and more uniform behaviour under UV irradiation, as shown by measured values and their standard deviations. Final condition of water drops is more strictly related to photo-induced hydrophilicity due to UV exposure than surface morphology of substrate and TiO2 coating itself. Capillary water absorption was evaluated through UNI EN 15801 standard test. Porous limestone has a high and rapid water absorption and, without UV light exposure, treated surfaces did not show evident differences due to TiO2 coatings compared with untreated case (Table 1). Water absorption depends exclusively by original physical and chemical properties of stone itself and not by TiO2 treatment, as shown by very high values of standard deviation regardless of the analyzed case, mostly due to irregularity of porous travertine surfaces and physical differences among samples. Most of absorption took place during the first part in contact with water and after 6 days of test procedure T1 showed no significant difference compared to the untreated case, while T2 surfaces increased capillary water absorption in a negligible way. To better assess wettability changes related to titanium dioxide presence, surface absorption of water sprayed by manual nebulizer was measured by weighing. Water absorption without UV light exposure is not related to coating presence or deposited TiO2 amount, since absorption values, their trend during time and the high standard deviations are mostly dependent on heterogeneous physical characteristics of the surfaces themselves, like porosity and roughness. Ultraviolet illumination activated photo-induced hydrophilism of titanium dioxide in an evident way, especially for T2 surfaces. Slight different values of water absorbed by untreated surfaces under UV irradiation did not derive from ultraviolet light exposure, as shown from still high standard deviation values due to travertine irregularity. As regard treated surfaces, water absorption values clearly decreased (Fig. 4). Differences between treatments were proportional to time of UV exposure: after 60 minutes, T1 average absorption value is 65 % of its respective unexposed case, while T2 value is 57 %, and these amounts are about 50 % of water absorbed by untreated samples (Table 2). Standard deviation values of treated cases were much lower, since wettability is strictly dependent on TiO2 photo-induced properties and treated samples showed a much more uniform behaviour less related to characteristics of porous limestone substrate. Obtained results showed in an evident way that photo-induced hydrophilicity of titania did not necessary lead to higher water absorption under UV exposure by treated substrates, since as expected water can create a film over TiO2 thin coating and solid surface sliding away by gravity without being absorbed. This behaviour seems to allow the use of TiO2 coatings on vertical or inclined stone surfaces, since increase of water absorption, a potential source of damage for stone surfaces, is avoided. Single layer and multilayer treatments showed moderate differences in wettability. It seems that deposition of successive TiO2 layers does not greatly influence photoinduced properties with respect to a single layer. The application of multiple layers of titania coating could lead to no benefits at least for short-medium times, only increasing costs and time of application. In order to evaluate photocatalytic self-cleaning effect, both de-pollution and dyedecolouration tests were carried out. Titania coatings clearly showed high efficiency in NO photo-degradation. During UV light exposure, single layer coating (T1) degraded about 35 % of NO concentration, while three layer treatment (T2) produced a 50 % reduction in NO concentration (Table 3). Untreated travertine showed no degradation at all as expected. De-pollution effect due to photoactivity of titanium dioxide lasts during all UV irradiation time and it ends as soon as UV lamp is turned off. Even if higher titania content can bring to greater phococatalytic activity, in this case an higher number of TiO2 layers does not mean a proportional higher degradation value, since just outer layer of coating is in contact with NO and can degrade the external polluting agents. Soiling photo-decolouration was evaluated by rhodamine B colorimetric test. From results is well evident that titania coatings can greatly accelerate the degradation process of organic dye: during first 4 hours of UV illumination most of decolouration of red stain took place (a* values drastically decreased) and differences between treated and untreated cases are very evident (Fig. 5). At the end of test procedure (26 hours of UV exposure) only multilayer coating T2 still had higher decolouration compared with untreated case, while T1 coating did not show any greater degradation value of stain in comparison with untreated case (Fig. 5). So the main function of TiO2 treatment is to make the degradation process faster. This behaviour can be very important in outdoor use, as long-term solar light exposure is not always available. Multilayer coating showed faster and higher decolouration of soiling, but the photo-degradation of stain is not proportional to greater TiO2 amount applied on travertine surface, since, as seen in NO degradation test, only the outer parts of the coating absorb or enter in contact with external agent (soiling). Conclusions In this paper the analysis of efficiency and compatibility of self-cleaning titania treatments with travertine surfaces was carried out. The use of self-cleaning surface treatments could be an interesting response to the problem of conservation of stone materials, especially in outdoor environment. Two different TiO2 coatings were realized on travertine surfaces by spray coating, obtaining a single layer and a multilayer treatments. In order to assess compatibility of titania coatings with porous stone surfaces in the field of Architectural Heritage, transparency of the coatings and the absence of harmful consequences of photo-induced hydrophilicity of TiO2 were evaluated. Analyzed titania coatings showed no changes on aesthetical properties of treated stones, concerning both colour and gloss. Photo-induced hydrophilicity of titania was well evident, but this increase in wettability did not necessary bring to higher water absorption or more contact with water by treated stone surfaces, since, as regard vertical or almost vertical surfaces, the water film due to hydrophilic effect of TiO2 can easily slide away by gravity. Finally, the efficiency of the coatings was evaluated by de-pollution and selfcleaning tests. Under UV illumination, both analyses showed very good results: titania coatings can easily degrade polluting substances and soiling. The deposited amount of TiO2 on stone surfaces does not seem to increase photoinduced properties in a proportional way, as just the outer layer of titania film seems to determine these properties coming in contact with UV-light and external materials to be degraded (polluting substances or deposited soiling). Applying multiple layers of titania could lead to no evident benefits at least for short-medium periods, just increasing costs. Long-term durability analyses by means of weatherometer and outdoor tests are currently under way to assess the properties of TiO2 coatings for better preservation of porous stone elements. Acknowledgements The authors would like to thank Professor Antonio Licciulli, Dr Sergio Franza and Daniela Diso (Salentec Srl) for the supply and application of TiO2 sol and their cooperation to this work. Tabelle, grafici e figure: Untreated T1 T2 Q5 (kg/m2) 0.104 ± 0.074 0.081 ± 0.068 0.118 ± 0.076 Q10 (kg/m2) 0.117 ± 0.083 0.092 ± 0.073 0.132 ± 0.082 Q30 (kg/m2) 0.134 ± 0.086 0.110 ± 0.078 0.156 ± 0.086 Q60 (kg/m2) 0.149 ± 0.085 0.132 ± 0.060 0.171 ± 0.088 Q1d (kg/m2) 0.228 ± 0.089 0.218 ± 0.051 0.239 ± 0.085 Q2d (kg/m2) 0.240 ± 0.094 0.230 ± 0.045 0.262 ± 0.087 Q6d (kg/m2) 0.270 ± 0.102 0.269 ± 0.050 0.300 ± 0.094 Table 1 – Average capillary water absorption values (Qi) as a function of time in contact with wet bedding layer (after 5, 10, 30, 60 minutes and 1, 2, 6 days). Untreated T1 T2 without UV under without UV under without UV Under light UV light light UV light light UV light 0.071 0.093 0.059 0.055 0.114 0.034 Q10 (kg/m2) ± 0.022 ± 0.043 ± 0.027 ± 0.021 ± 0.043 ± 0.004 0.167 0.225 0.130 0.129 0.200 0.080 Q30 (kg/m2) ± 0.046 ± 0.095 ± 0.064 ± 0.015 ± 0.085 ± 0.021 0.311 0.220 0.144 0.304 0.172 2 0.269 Q60 (kg/m ) ± 0.132 ± 0.153 ± 0.108 ± 0.027 ± 0.133 ± 0.016 Table 2 – Surface water absorption average values (± standard deviation) during time (after 10, 30 and 60 minutes of water spraying), without UV irradiation and under UV light exposure. T1 T2 Before UV exposure (ppm) 0.61 0.61 Under UV exposure (ppm) 0.40 0.30 After UV exposure (ppm) 0.61 0.61 Table 3 – Average nitrogen oxide concentration (ppm) before, during and after UV irradiation. Fig. 1 – SEM images (magnification 2000x) and EDS results of treated travertine surfaces. Fig. 2 – Maintenance of aesthetical properties: colour changes due to TiO2 coatings. Fig. 3 – Average static contact angle of water drops under UV illumination. Fig. 4 – Average absorption of nebulised water: differences between unexposed and UV-exposed cases. Fig. 5 – Degradation of organic dye (rhodamine B) on travertine under UV light. References: Bondioli F, Taurino R, Ferrari AM, Functionalization of ceramic tile surface by sol-gel technique, Journal of Colloid and Interface Science 334 (2009) 195-201 Chen J, Poon C, Photocatalytic construction and building materials: From fundamentals to applications, Building and Environment 44 (2009) 1899-1906 Diamanti MV, Ormellese M, Pedeferri MP, Characterization of photocatalytic and superhydrophilic properties of mortars containing titanium dioxide, Cement and Concrete Research 38 (2008) 1349-1353 Fujishima A, Honda K, Electrochemical Photolysis of Water at a Semiconductor Electrode, Nature 238 (1972) 37-38 Fujishima A, Rao TN, Tryk DA, TiO2 photocatalysts and diamond electrodes, Electrochimica Acta 45 (2000) 4683-4690 Luvidi L, Laguzzi G, Gallese F, Mecchi AM, Sidoti G, Application of TiO2 based coatings on stone surfaces of interest in the field of Cultural Heritage, Proceedings of 4th International congress Science and technology for the Safeguard of Cultural Heritage in the Mediterranean Basin, Cairo, December 6-8, Vol. II (2010) 495-500 Potenza G, Licciulli A, Diso D, Franza S, Calia A, Lettieri M, Ciccarella G, Surface engineering on natural stone through TiO2 photocatalytic coating, Proceedings of International RILEM Symposium on Photocatalysis, Environment and Construction Materials, Florence, October 8-9, (2007) 315-322 Quagliarini E, Bondioli F, Goffredo GB, Licciulli A, Munafò P, Smart surfaces for Architectural Heritage: preliminary results about the application of TiO2-based coatings on travertine, Journal of Cultural Heritage, in press, doi:10.1016/j.culher.2011.10.002 Quagliarini E, Bondioli F, Goffredo GB, Licciulli A, Munafò P, Self-cleaning materials on Architectural Heritage: compatibility of photo-induced hydrophilicity of TiO2 coatings on stone surfaces, Journal of Cultural Heritage, in press, doi:10.1016/j.culher.2012.02.006 Zhao X, Zhao Q, Yu J, Liu B, Development of multifunctional photoactive selfcleaning glass, Journal of Non-Crystalline Solids 354 (2008) 1424-1430