TIO2 PHOTOCATALYSIS FOR IMPROVING THE AIR QUALITY: FROM MOLECULES, TO BUILDING MATERIALS DEVELOPMENT

General abstract 1. Introduction During these last years, the innovation and development processes lead pollution to its highest level; the air pollution is one the most prominent and dangerous form of it. Causes are several, from fuel combustion to factories activity, which increase the level of organic molecules and nitrogen or sulfur oxides in atmosphere (WHO Global Urban Ambient Air Pollution Database, update 2016). Unfortunately, effects are more than evident: from the global warming, to the acid rains, from the sudden climate changes, to the increase of diseases such as asthma and lung cancer (Ambient (outdoor) air quality and health, WHO). Outdoor air pollution is the major environmental health problem affecting everyone in developed and developing countries alike, however, unlike one might usually think, indoor levels of organic pollutants are often higher than outdoor (Viegi et al., 2004). The problem is even more important because people live mainly indoors, constantly exposed to all the pollutants present in these close environments (Chen et al., 2016; Allen et al., 2016). For this reason, demands to improve the air quality situation have been largely extended, finding new strategies for waste reduction or for the oxidation and degradation of pollutants (Ambient (outdoor) air quality and health, WHO). Among several processes, considering that very important factors are saving energy and reducing emissions, photocatalysis has been exploited as very suitable technique to reduce pollution. In a photocatalytic reaction (eq. 1), a semiconductor material, the photocatalyst, is activated by light and, thanks to the formation of some electron-hole couples between his valence and conduction bands, it is able to reduce or oxidize molecules that adsorb on his surface (J.M. Herrmann, 2005). In heterogeneous photocatalysis, the reaction implies the previous formation of an interface between the semiconductor and the reactants of the reaction (K. Demeestere et al., 2007; M. Schiavello, 1997). (Ox1)ads + (Red2)ads Red1 + Ox2 (1) Among a large variety of semiconductor materials, which are mainly metal oxides, only few of them are considered to be applicable photocatalysts, in relation with their specific photocatalytic properties. Titanium dioxide (TiO2) induced photocatalysis is an example of AOP processes and it has been demonstrated its efficiency in the decomposition of various organic contaminants. TiO2 is a very well known and well-researched material due to the stability of its chemical structure, biocompatibility, physical, optical, and electrical properties (M. Serpone et al., 1989). The crystalline forms of TiO2 are anatase, rutile and brookite (A. Linsebigler et al., 1995). In general, TiO2 is preferred in anatase form because of its high photocatalytic activity, however the major drawbacks of TiO2-based photocatalysts is related to the rapid charge recombination of the electron−hole pairs, and the wide band gap, which restricts light absorption to only ultraviolet region (wavelength <390 nm), restraining the practical applications of TiO2-based photocatalysts under solar light or visible light. TiO2-based photocatalysts are used for a variety of applications such as degradation of volatile organic compounds (VOCs) and decomposition of nitrogen pollutants (NOx) or also organic dyes, like Methylene Blue (K. Demeestere et al., 2007; P.K.J. Robertson et al., 2005). When TiO2 is irradiated with energies equal to or higher than its band gap (>3.0 eV), electrons are excited from the valence band into the conduction band, leading to excited electrons in the conduction band and positive holes in the valence band. This fundamental process can be expressed by the following reaction equations (eq. 2): TiO2 + hv → e- + h+ (2) As electrons have a reducing potential, holes can oxidize water and lead to the formation of more oxidant species such as hydroxyl radicals, able to oxidize organics. As mentioned above, the field of practical applications of TiO2-based photocatalysts becomes less expanded under solar light or visible light. In this sense, different strategies have been developed (X. Li et al., 2011; S. Afzal et al., 2013; S. Wu et al., 2013; Y. Cho et al., 2001), starting from the chemical modification of TiO2 lattice using non-metals, particularly carbon, nitrogen and sulfur (S. Khan et al., 2002; J. Gole et al., 2003; T. Umebayashi et al., 2003). The presence of metal nanoparticles on TiO2 surface can promote charge transfer process in the composite systems (N. Chandrasekharan et al., 2000; A. Dawson et al., 2001), because of the electron injection that occurs from the nanosurface to the conduction band of TiO2 and the metal particle. In recent years, formation of photocatalytic heterostructures based on TiO2 with other semiconductor/noble metal has emerged as an important strategy to increase the separation of charge carriers and suppress the recombination rate of photoinduced electron−hole pairs, resulting in improved photocatalytic efficiency (F.X. Xiao et al., 2012; B. Liu et al., 2011; V. Etacheri et al., 2013; V. Etacheri et al., 2010; V. Etacheri et al., 2012; Y. Wang et al., 2013). Aims of the work The aims of the work regard different points of the TiO2 study and improvement. Starting from the choice of the best commercial powder of TiO2 to replace the titanium oxide nano-powders, through its modification to make it useful in visible light, until the application on ceramic supports to prepare building materials appropriate for outdoor and indoor pollution abatement, to improve the air quality and the quality of our life as well. The purposes of this research project can be therefore summarized as follows: • Study in depth the real potential of a micrometer TiO2, finding the best candidate among several commercial samples to obtain benefits such as economic saving, safety, ease in product handling in industry; • Improve the use and application of micro-TiO2 on building materials to optimize their performances in pollution abatement, with particular attention to realistic settings; in this sense, find innovative methods to test materials and assesses the photocatalytic potentials; • Make micro-TiO2 active under visible light, modifying it with noble metals and in particular combining the process with the use of high-energy ultrasound (US). 2. Experimental details 2.1 Commercial samples of TiO2 In this work, the starting point was the study of different commercial powders of TiO2. Five commercial TiO2 materials by Kronos, Hundsman, Sachtleben (two different powders) and Cristal (respectively quoted with the B–E letters) have been selected. They are all available in the market as pigmentary powders, and they have the following key features: pure anatase phase, uncoated surface, undoped material, not sold as photocatalytic material. P25 by Evonik is the nanometric commercial TiO2 reference for photocatalytic applications, and it is the most used and studied. All commercial powders were used as received without any further treatment or activation process (C.L. Bianchi et al., 2015). The crystallographic phase composition have been valued by XRD patterns and all the samples are pure anatase, except for P25. Crystallite size of 1077 by Kronos endorses its micro-sized nature, always connects to a low surface area. As expected, P25 is a nano-sized powder. Studying more thoroughly the morphological characteristics particularly by TEM analysis, it can be notice that the reference P25 powder is made up of well-crystallized particles of rather roundish shape, closely packed and with an average size of 20–30 nm. XPS results give information about the surface states of TiO2 and there are not differences among all the present samples concerning binding energy (BE). Even the band gap values, evaluated by means of diffuse reflectance UV–Vis analysis, do not exhibit large differences among the various samples. 2.2 Selected pollutants Photocatalysts, whether commercial as such or modified as explained in the next sections, have been tested on VOCs molecules referring to indoor pollution, and on nitrogen oxides (NOx) in reference to outdoor pollution. Some reagents are liquid, other are gaseous and stored in cylinders under pressure. All substances were used as purchased without any particular pre-treatments or purification. 2.3 Photocatalytic reactors VOCs degradation Photocatalytic degradations were conducted in a Pyrex glass cylindrical reactor with diameter of 200 mm and effective volume of 5 L. The catalyst in powder form has been deposited on a flat glass sheet (100cm2) as thin film, from a suspension in 2-propanol. The catalyst amount used in each tests was 0.05 g. The atmosphere in the reactor was obtained by mixing hot chromatographic air humidified at 40%, and a fixed amount of volatilized pollutant, in order to avoid condensation. Photon sources were provided by a 500 W iron halogenide lamp (Jelosil, model HG 500) emitting in the 315–400nm wavelength range (UV-A) at 30Wm-2, or by a LED (MW mean well, 350 mA rated current, 9–48 V DC voltage range, 16.8 W rated power) with an emission between 400 and 700 nm. The actual concentrations of pollutant in the reactor were determined directly by micro-GC sampling or by Proton transfer reaction mass spectrometry (PTR-MS) (detailed description is reported in the next sections). Nitrogen oxides degradation ➢ NOx photocatalytic degradations were conducted in a Pyrex glass cylindrical reactor with an effective volume of 20 L in batch mode. The catalyst in the form of powder has been deposited from 2-propanol suspension on a flat glass sheet (40cm2), and the amount used in each tests was 0.05 g. The gaseous mixture in the reactor was obtained by mixing NOx (mixture of NO and NO2 in air) with air humidified at 40%. The initial concentrations of NOx in the reactor were 1000 ppb in order to follow the same pollutant concentration requested by the ISO 22197-1 rules (www.iso.org) and 200 ppb that is very close to the alert threshold set by the EU Directive 2008/50/CE for NO2 (http://eur-lex.europa.eu). Photon source was provided by a 500 W iron halogenide lamp (Jelosil, model HG 500) emitting in the 320–400 nm wavelength range (UV-A). The specific UV power on the surface of the samples was 10 Wm-2. The concentration of pollutants in the reactor was determined directly by chemiluminescence (Teledyne, Mod. 200E). ➢ The continuous flow reactor has been used only for testing photocatalytic building materials; it has got walls of 10 mm in thickness, and an internal size of 625 × 625 × 115 mm3, with four inlets and one opposite outlet and can house a sample of 600 × 600 × 10 mm3. It is equipped with a thermo-hygrometer model HT- 3006A to measure the temperature and humidity during the tests. The relative humidity inside the reactor is maintained constant around a value between 40 and 50%. The experiments were carried out either using UV lamps (UV-A region, 20 Wm−2) or using sunlight from July to September. The degradation was performed at different initial NOx concentrations ranging from 100 ± 10 ppb to 200 ± 10 ppb, at room temperature and working with total gas flow rates of 140 and 180 NL h−1. Even for these tests, the concentration values were chosen in order to work closely to the limit values reported on Directive 2008/50/EC, in particular, 106 ppb (equal to 200 μg m−3) and 213 ppb (400 μg m−3, alert threshold). The duration of each continuous run was set at 6, 12 or 24 h. The final design of the reactor was selected among several possibilities by considering the good homogeneity of the reactant in the gas phase and a contact between the reactant and the photocatalytic material that effectively reproduce the real working conditions. 2.4 Samples characterization The morphology of TiO2 in form of powder, both commercial and synthesized or modified, was inspected by means of high-resolution transmission microscopy (HR-TEM) (Jeol JEM 3010 instrument, equipped with LaB6 filament and operating at 300 kV), and the surface area of all the catalysts was determined by conventional N2 adsorption (BET) at 77 K using a Sorptometer (Costech Mod. 1042). XRD spectra were collected using a PW 3830/3020 X’ Pert Diffractometer from PANalytical working Bragg-Brentano, using the Cu Kα1 radiation (k = 1.5406 Å). X-ray photoelectron spectra (XPS) were taken in an M-probe apparatus (Surface Science Instruments). Diffuse reflectance spectroscopy (DRS) of the ground powders was performed with a Thermo Scientific Evolution 600 spectrophotometer, equipped with a diffuse reflectance accessory Praying-Mantis sampling kit (Harrick Scientific Products, USA). A Spectralon1 disk was used as reference material, and the experimental absorption versus lambda plot was elaborated using the Kubelka–Munk function. Absorption/transmission IR spectra were obtained on a Perkin-Elmer FT-IR System 2000 spectrophotometer equipped with a Hg–Cd–Te cryo-detector. Particularly for metals-modified TiO2, ICP/OES analysis has been performed using a Perkin Elmer Optima 8300 instrument. HR-SEM-EDX analysis was performed particularly on photocatalytic building materials (tiles) (Field Emission Gun Electron Scanning Microscopy LEO 1525, metallization with Cr. Elemental composition was determined using Bruker Quantax EDS). The surface wettability was evaluated by static contact angle (CA) measurements performed with an OCA20 instrument (DataPhysics Co., Germany) equipped with a CCD camera and a 500μL-Hamilton syringe to dispense liquid droplets. 2.5 Metal NPs modified TiO2: synthesis procedure To obtain the surface modification of the commercial powder of TiO2, in this work we performed a synthesis by means of high-energy US. Procedure steps are slightly different from a metal to another, but in general they follow the same scheme, described in the next lines. The precursor materials are organic or inorganic salts of different metals, selected in accordance with characteristics that will be detailed in the following chapters. For the US generation, we used A Bandelin SONOPLUS HD 3200 utilizing a 200W U/S generator and a sonication extension horn of 13mm diameter. Generally, the metal precursor and the commercial powder of TiO2 have been put together in a 100 ml glass flask, and they have been solved with the preferred solvent (aqueous or organic). The solution is then sonicated at constant temperature with a specific amplitude and intensity (Wcm-2). At the end the solution is centrifuged many times to remove all the solvent and the final powder is washed before evaporation and/or calcination steps. 2.6 Building materials production: ACTIVE® photocatalytic tiles (by GranitiFiandre S.p.A) Airless spray classic preparation Industrial porcelain grés tiles are manufactured under high pressure by dry-pressing fine processed ceramic raw materials with large proportions of quartz, feldspar, and other fluxes and finally fired at high temperatures (1200–1300°C) in a kiln. To obtain photoactive porcelain grés tiles, they were subsequently covered by airless spray with a mixture of micro-TiO2 (specifically, 1077 by Kronos has been selected as best commercial powder) mixed with an aqueous suspension of a commercial SiO2-based compound (process developed by GranitiFiandre S.p.A, patent n. EP2443076). At the end of the preparation procedure, tiles were fired at high temperature (min 680°C) for 80 min. Finally, the powder present at the sample surface and not completely stuck was brushed and removed. Digital Printing technology: DigitaLife Project The digital printing technology is based on suitably designed print heads using a tailored solvent-based ink, micro-sized TiO2 and additives (process developed by GranitiFiandre S.p.A, DigitaLife project). Specific and more detailed information will be given in the various chapters dedicated. 3. Results and discussion 3.1 Characterization of the TiO2 powders and materials Among the starting selected commercial samples, 1077 by Kronos has been chosen as best micrometer candidate to replace the nanometric reference P25. For this reason, all the results that will be presented in this thesis refers to P25 or 1077 as commercial references, nano and micro-sized respectively, and, all the structural modification studies, synthesis, tests, have been performed using 1077 as TiO2 support on which make changes. HR-TEM images confirm the nanometric nature of P25 and the micrometric dimension of the TiO2 particles of 1077 commercial sample. Results are absolutely completely in line with the results about the surface area (gm-2), which is very low in case of bigger TiO2 particles. From the same characterization analysis performed on the various modified samples, it is shown that this structure composition has not been changed by thermic treatments (calcination steps) or modification steps of the original sample with metals, or by classical impregnation methods either by using ultrasound as will be described in the dedicated chapters. XRD spectra give particularly information about the crystallographic phase composition. As presented in the table above, 1077 consists completely of anatase, which is a very good feature in term of photocatalytic activity. Any modification steps have not altered even this composition. All the samples presented in this work, except P25, consist of anatase. Useful in term of photocatalytic activity is also the distribution on the TiO2 surface of the OH groups, which are measurable in relation to the total oxygen (OTOT), particularly by means of XPS analysis; 1077 by Kronos presents a higher OH/OTOT ratio (0.14 and 0.32 for P25 and 1077 respectively) than P25, and this value, ascertainable even more accurately using the IR spectra, increases modifying 1077 TiO2 by metals. When TiO2 is decorated on surface with noble metal or metal-oxides nanoparticles (for every modified-metals-TiO2 sample presented in this work, we have usually a mixture of metal and metal-oxides NPs) the UV-VIS spectra show that the absorption shifts to the visible wavelengths, more or less depending on the metal species and its amount deposited on TiO2. UV-VIS spectra collected on the references, P25 and 1077 respectively, confirm the slight absorption in the visible of P25, because of the presence of rutile, and the absence of absorption in visible of 1077 by Kronos, which can be activated only by UV irradiation, even because its band gap, typical of anatase TiO2 (3.2 eV). Referring to the surface area, presented above for the references commercial powders, deposition of NPs on TiO2 surface can have slights effects on the final surface area, which increases, even if the value is very small and almost negligible. Finally, SEM characterization is very useful to study the ceramic surfaces when TiO2 is deposited in them. In particular, the main points in case of the two photoactive grés ceramic tiles samples are a) the fact that 1077 TiO2 does not change its nature so remains anatase and micrometer; b) changing the deposition method the distribution of the TiO2 particles completely changes. 3.2 Pollutants photodegradation: Nitrogen Oxides The use of a pigmentary powder of TiO2 as 1077 by Kronos, consisting in micrometer particles with a lower surface area, is absolutely effective in case of NOx abatement. The comparison between 1077 and P25 (as reported in Table 4) shows that the degradation percentage obtained after 120 minutes of photocatalytic reaction is very slight and almost negligible, if we consider the advantages in term of economic saving (2$/kg for P25 vs. 0.45$/Kg for 1077) and safety. The main reasons why 1077 shows to be a very good candidate between various pigmentary and commercial TiO2 in photocatalysis, are firstly its phase composition, i.e. anatase, without rutile. Moreover we have to consider that 1077 surface present a wide population of OH groups, which both for the adsorption of pollutant molecules and for them degradation are crucial. Therefore, especially for NOx abatement, micro-TiO2 as it is proves to be efficient. Thus, a modification of the material with metals, which increase the final cost, is unnecessary. 1077 however presents a very low activity if irradiated only by visible wavelengths, as confirmed by the UV-VIS spectra, which do not show absorption peaks after 400nm (visible spectrum from 400nm to 700nm). In this case, the surface modification of TiO2 with metal NPs is a key factor to have effective samples in nitrogen oxides abatement. Preparation methods are different and they will be deeply described in the dedicated chapters. To summarize the most important results, as can be seen from the presented graphs (Fig. 4-6), the presence of silver nanoparticles clearly improve the photocatalytic activity of the sample, and the key factor related to a better NOx degradation are particularly the metal-NPs amount and dimension. Finally, after showing that 1077 by Kronos is effective, it has been deposited on ceramic grés tiles as already mentioned. The borderline between samples Z23 and S24 is the method with which the deposition is obtained. Digital printing technology leads to a better and more uniform distribution of the TiO2 powder, as well as to a lower loss of it during the process, with the final results clear presented in the kinetics graphs (see Fig. 4). 3.3 Pollutants degradation: VOCs Commercial powders of TiO2 have been already exploited in VOCs photodegradation (C.L. Bianchi et al., 2014; S.B. Kim et al., 2002; G.M. Zuo et al., 2006), showing very good results against various molecules (see tab.5). However, and it is more evidently than in nitrogen oxides photodegradation, 1077 by Kronos is slightly less effective mainly due to its lower surface area. Moreover, the oxidation of more complex molecules, as in case of toluene that is an aromatic compound, both P25 or 1077 are not able to reach the complete degradation of it after 6 hours of UV irradiation. For this reason, even for this kind of application, the modification and improvement of the TiO2 photocatalyst is necessary. The presence of metals and metal oxides on the TiO2 influences the electrons-holes separation, the number of available electrons, the band gap value and the organic molecule adsorption respectively (V.E. Henrich, 1994; H. Al-Abadleh and Grassian, 2003). Among different metals, with properties that will be deeply explained in the dedicated chapters, silver seems to be the best in term of organics oxidation, showing a consistent improvement in the degradation reaction of toluene as reported in fig. 7. Metals-surface decoration is essential under visible light. Anatase-TiO2 with a band gap of 3.2 eV (C. Dette et al., 2014) is not photocatalytically active. Thus, the presence of metals on its surface is needful. The most important factors that influence the final result are i) metal species and the nature of the species deposited on the TiO2 surface; ii) metal amount; iii) metal NPs average dimension; iv) metal NPs distribution. In this sense, the synthesis method is very important because can change these latter parameters, and the use of US during the TiO2 decoration step as reported in (M. Stucchi et al., 2016; M. Stucchi et al., 2014) are very important. US do not change the morphology of TiO2, but they are crucial for the formation of metal and metal oxides nanoparticles as well as for their good distribution on the semiconductor surface, as well as, particularly in case of silver, to obtain bigger spherical particles. The number of silver nanoparticles distributed onto TiO2 surface affects the final photoactivity, and finally a higher number of NPs leads to a better caption of the visible light electrons. 4. Conclusion As a conclusion to this work, some final remarks can be claimed: ➢ Comparing the photocatalytic performances of various commercial TiO2 powders with those of the reference P25 system, in the degradation of different VOCs, such as acetone, toluene, or ethanol, or NOx, different performances have been achieved. The different performances achieved by the various materials are representative of the different physico-chemical features of them. With the main aim to replace the nanosized P25 with a micrometer powder of TiO2, 1077 by Kronos shows the best performances. Thus, the present research indicates first that also micro-sized TiO2 powders, of commercial origin and normally employed as pigments, are very promising materials to be used in the photocatalytic degradation of VOCs and NOx, which would help limiting the risks for human health deriving from the use of nanoparticles. ➢ Then, comparing specifically the photocatalytic performance of P25 and 1077 by Kronos, it is shown that: the adsorption of the pollutant molecule at the semiconductor surface is important to promote the photocatalytic reaction, and in this sense, FTIR analysis of the OH stretching region demonstrated the presence of a good amount of Ti-OH-Ti bridged species on the surface of the micro-sized TiO2, which plays a key role in driving the photocatalytic activity. ➢ Through the innovations introduced by sonochemistry, it was possible to obtain a new type of surface decoration of the pigmentary micro-TiO2, proving that this modification method can improve the photocatalytic activity of the material, in particular under the visible light, where pure TiO2 is not an effective photocatalyst. Indeed, when TiO2 is irradiated by visible wavelengths only, the photocatalytic activity is completely lost and particularly, the deposition of both metal or metal oxides nanoparticles, can positively affect its activation under visible light. ➢ In this research project, the features of Ag, Cu and Mn, have been respectively studied, in particular focusing the attention of the preparation method and the metal amount deposited on the TiO2 surface. It is possible to summarize that: i) high energy US is a facile and fast method to obtain the surface decoration, even when the support is commercial and micrometric; US do not change the morphology of TiO2, but they are crucial for the formation of metal and metal oxides nanoparticles as well as for their good distribution on the semiconductor surface; ii) copper proves to be a good candidate, and its amount is crucial, because as a higher amount of copper both increases the absorption of the visible light and improves the electron-hole separation, over a certain amount of metal the performance decreases, because of the excessive coverage of the active sites of TiO2; iii) Ag shows even better properties in term of TiO2 activation under visible light. Even in this case US do not change the morphology of the micrometric support, and they are important to obtain bigger spherical particles of silver as well. Moreover, the number of silver nanoparticles distributed onto TiO2 surface affects the final photoactivity, and the correlation is linear, because an higher number of NPs leads to a better caption of the visible light electrons first, as well as they reduce the electron-hole recombination, acting as electron traps. Suggestions for the continuation of the work concern the development of production processes to apply the new metal-modified TiO2 powders on ceramic materials, to obtains building products active and effective particularly in indoor environment. Secondly, concerning the use of ultrasound in metal NPs synthesis and their application on TiO2, study further its effects in the pollutants degradation kinetics, as well as in the photocatalytic reactors rheology. Finally, investigate further the degradation of organic molecules with different effects from air pollution: ethylene is harmless for human health, but it is the fruit-ripening hormone. Thus TiO2-materials can be studied for applications regarding food storage and preservation.