Solar Energy Materials and Solar Cells 188 (2018) 127–139 Contents lists available at ScienceDirect Solar Energy Materials and Solar Cells journal homepage: www.elsevier.com/locate/solmat Non lithographic block copolymer directed self-assembled and plasma treated self-cleaning transparent coating for photovoltaic modules and other solar energy devices T Deepanjana Adaka, Sugato Ghosha, Poulomi Chakrabortyc, K.M.K. Srivatsad, Anup Mondalb, ⁎ Hiranmay Sahaa, Rabibrata Mukherjeec, Raghunath Bhattacharyyaa, a Centre of Excellence for Green Energy and Sensor Systems, Indian Institute of Engineering Science and Technology, Shibpur, Howrah 711103, West Bengal, India Department of Chemistry, Indian Institute of Engineering Science and Technology, Shibpur, Howrah 711103, West Bengal, India Instability and Soft Patterning Laboratory, Department of Chemical Engineering, Indian Institute of Technology, Kharagpur, West Bengal 721302, India d Physics of Energy Harvesting Division, CSIR-National Physical Laboratory, Dr. K.S. Krishnan Marg, New Delhi 110012, India b c A R T I C LE I N FO A B S T R A C T Keywords: Antireflective Self-assembled Self-cleaning Photo-active Plasma-treatment Non-lithographic Through a combination of sol-gel based self-assembly and plasma based approach we have developed highly transparent, self-ordered, superhydrophilic and photoactive TiO2 thin film coatings. TiO2 sol used for such coatings comprises a block copolymer which functions as a structure directing agent. This structure directing agent aid to formation of regular pores in the TiO2 thin film, thereby, remarkably reducing the refractive index values (~ 1.31) and enhancing the transparency (4% antireflection gain) of the coatings. Further, such porous TiO2 coatings show an excellent ability to photo-decompose organic pollutants, due to the photocatalytic ability of such metal oxide semiconductor. Enhancement in the photocatalytic activity has been obtained by porous surface created using a block copolymer and shifting the band gap energy by incorporating nitrogen so as to utilize part of the visible region of the solar spectrum for photocatalysis. An optimum condition is achieved by varying the RF self-bias potential and time of plasma treatment. Nitrogen plasma treatment, in addition to enhancing the photocatalytic activity of TiO2 is also found to enhance the mechanical stability and hydrophilicity, without hampering the optical transmission of coatings. Such coatings are also found to exhibit superhydrophilicity with water contact angle (WCA) < 5° under optimized condition. Thus, the coatings developed, qualify as a suitable candidate to be applied on solar PV panel and other energy devices. Treatment with nitrogen plasma extends the photocatalytic activity towards visible region of the spectrum and also ensures the mechanical stability of the otherwise porous network. 1. Introduction With the ever rising technological advancement to leverage solar energy in the modern era, maintaining optimum performance of such energy devices has become an important issue. The major solar energy applications are in the form of photovoltaic (PV) module to generate electric power and concentrated solar thermal power (CSP). The dust fouling has an extremely detrimental effect on PV panels and it varies depending on area and its environmental conditions [1–3]. It has been reported that in dry and lower rainfall areas like Saudi Arabia that loss may range as high as 26–40% and the transmission of glass cover on PV panels reduced to 70% because of dust in summer [4,5]. Under USA SunShot CSP program, Oak Ridge National Laboratory has developed optically transparent superhydrophobic materials and coatings based ⁎ on nanostructured silica surfaces that can address the soiling and maintenance issues of CSP systems by maintaining optimized adhesion, optical transmission/reflection, water and dirt repellent properties [6]. Superhydrophobic surfaces comprises of micro-/nanostructures and a low-surface energy self assembled monolayer (SAM). In order to achieve self-cleaning, the surface must have low adhesion and high water contact angle. These surfaces mimic the self-cleaning mechanism of the lotus leaf [7]. It has, thus, been observed over a period of time that superhydrophobic coatings gradually begin to fail and the performance starts to reduce, exhibiting poor self-cleaning performance than uncoated glass [8,9]. On the other hand, more durable superhydrophilic surfaces can be achieved through surface structuring, depositing nanoparticle films, or photo-induced hydrophilicity (PIH) of semiconductor films [10–12]. As the liquid film on the surface spreads Corresponding author. E-mail address: drraghunath@gmail.com (R. Bhattacharyya). https://doi.org/10.1016/j.solmat.2018.08.011 Received 17 April 2018; Received in revised form 16 August 2018; Accepted 17 August 2018 Available online 06 September 2018 0927-0248/ © 2018 Elsevier B.V. All rights reserved. Solar Energy Materials and Solar Cells 188 (2018) 127–139 D. Adak et al. triblock copolymer Pluronic F127 in combination with metal oxide based sol-gel chemistry. In this approach the block copolymer acts as a structure directing agent to impart required porosity needed to achieve enhanced optical transmission. In order to achieve the required antireflection behavior in such coatings there must be destructive interference of light reflecting off the two (or more) film surfaces which can be typically achieved by proper tuning of thickness and refractive index of constituent layers. Two conditions have to be fulfilled for a single layer ARC: (1) For a given wavelength (λ) and angle of incidence, the required optical thickness of the ARC must be λ/4, and (2) the amplitude matching of the two reflected beams requires an effective refractive index of the ARC (ηAR), which is the square-root of that of the optical substrate [30]. Most common transparent substrate material such as glass has a refractive index of around 1.5, thus requiring ηAR~1.22. Low iron toughened glass which has also refractive index close to common glass substrate is used to cover solar panels. A further decrease in the refractive index can only be achieved by the introduction of porosity on the sub-wavelength length scale. Most of the research on sol-gel based approaches emphasize on porous silicabased material, offering tunable refractive index and thickness with an appreciable adhesion to glass surface. In numerous occasion it has been found that the pores have a tendency to adsorb pollutants under outdoor environment which significantly affect the optical properties of such coatings, thereby, resulting in poor antireflection behavior. Moreover, such coatings are also mechanically fragile. Absorption of organic contaminants from the ambient atmosphere results in the deterioration of photovoltaic module efficiency and affects the optimum performance and its long-term usability, requiring more frequent mechanical cleaning. Thus, for long term sustainability of PV modules photoactive metal oxide semiconductor i.e. TiO2 is preferred as a coating to degrade the adsorbed hydrocarbons [31]. However it needs to be suitably tailored to achieve optimum AR behavior for instance by paring with silica as a mixed composition or can be used as a bottom layer coating [10,32,33]. Another elegant approach to overcome this problem is using block copolymer as a structure directing agent to prepare porous TiO2 coatings. In addition to having desirable anti-reflecting and photoactive properties such coatings need also to show either superhydrophilic or superhydrophobic nature with WCA < 5° or > 150°, as the case may be. Faustini et al. reported double layered anti-reflecting coating using block copolymer assisted patterned TiO2 with refractive index as low as 1.756 at 700 nm [34]. Later on Guldin et al. reported self-cleaning and low cost antireflecting coating by self-assembly of block copolymer in combination with sol-gel chemistry to obtain TiO2 nanocrystals with regular porosity [31]. However, in most of the cases the block copolymer being used is very expensive restricting their usage to the surfaces with small area. Thus, in order to explore the possibility of block copolymer as a structure directing agent we explored various less expensive block copolymers and choose Pluronic F127 for our experiments. This triblock copolymer is not only cost efficient but also compatible with the solvent used to synthesize TiO2 coatings by sol-gel chemistry. It is noted that Miao et al. have demonstrated application of porous TiO2 coating synthesized using Pluronic F127 block copolymer in the form of double layered SiO2-TiO2 coating on the solar glass cover [35]. Although, with such bilayer systems they have reported an average of 3.4% gain in transmission, the self cleaning activity is noticeable only on prolong exposure to UV light irradiation. By taking motivation from this work we tried to develop single layer structured TiO2 thin films with regular porosity and further treated it under nitrogen plasma to shift the band gap of otherwise wide band gap metal oxide semiconductor to make use of some of the visible part of the solar spectrum as well. However, due to porosity of these functionalized templated mesoporous TiO2 thin films, optical and structural characterization is a major challenge. In order to overcome this problem Ortel et al. has proposed an approach to determine the porosity of thin films by means of electron probe microanalysis (EPMA) either by rapidly, it can dislodge contaminants and remove them from the surface. Self-cleaning superhydrophilic coatings have been found to outperform superhydrophobic coatings at both the lab and industrial environment [9,13]. Most of the superhydrophilic coating is found to consist of TiO2, a photocatalyst, capable of breaking down organic contaminant when exposed to light [14,15]. Some commercial hydrophilic self-cleaning coatings have also emerged in the recent past. These are mainly chemical based coatings that are coated on ceramic materials using spray technique. A well-known Japanese manufacturer, Sketch Co. has come up with nano-coating materials that provide coated substrates with a variety of functions such as anti-static, superhydrophilic, with an appreciable degree of hardness, weather resistance property and high optical transparency. Besides, there are some commercially available self-cleaning glass coatings such as Pilkington Activ (TiO2 based hydrophilic coating) [9]. All such self-cleaning coating glass surfaces, other than possessing desired mechanical property, should also not be compromised in its high optical transparency. Other successful coating recipes, often using sol-gel process, are being used by solar panel manufacturers like Yingli Solar (CleanARC). Sprayable formulations like TitanShield (TSG80-01HD), Nanomagic (Magic Solar CoatSiO2, etc. are also available. Similarly, Sentech (China) offer commercially Self-cleaning features in their monocrystalline Solar panels. Sandia National Laboratories, USA have developed a Sol-gel based coating recipe which has been licensed to Samsung for their display applications. Our aim is to come out with a chemistry based (Sol-gel & Plasma Chemistry) highly transparent self-cleaning coating recipe that can be applied on the top side of a low iron tempered glass used in fabricating Si solar PV panels [16]. Their application again is not limited to Solar Glass cover and cost of ownership of the basic equipment will be comparatively low. Solar glass covers consists of low iron toughened glass to sustain extreme weather conditions like hail storm, dust storm, high temperature and humidity [17]. To achieve large-scale selfcleaning coatings on such energy systems several techniques have been developed and implemented that uses chemistry-based approaches such as spray coating, dip coating and aerosol deposition [1,18–20]. Other wet chemical methods which have been widely studied include layerby-layer assembly, other variants of sol-gel process and nanoparticle coatings using suitable binder [21–24]. Sol-gel coatings were studied and applied since long in various fields such as AR coatings, optical filters, biomedical instruments, textile and ceramic industries [25]. Again one of the very promising techniques for applying selfcleaning coating is a gaseous plasma based approach which includes pretreatment of substrate under oxygen, argon and other gases in plasma state, post deposition plasma treatment, plasma based etching of substrates using different fluorinated hydrocarbons and plasma assisted aerosol vapor deposition process [26,27]. By integrating both the sol-gel and plasma-based techniques it has been possible to develop coatings which exhibit unique optical and surface properties. The basic aim is to prepare a self-cleaning coating which can either be superhydrophilic or superhydrophobic in nature with appreciable optical transmission that will either maintain or enhance the transparency of glass cover. In addition to enhancing antireflection behavior using such combined processing we could make use of photocatalytic activity of TiO2 to degrade organic contaminants even in the presence of visible light by incorporating N atoms through plasma treatment, thereby, reducing the energy band gap of TiO2 thin film. It may be noted that, unlike other approaches using a bilayer of SiO2 and TiO2, in the present study, we report a single layer structured titania film, which has been found to satisfactorily meet the desirable functional requirements of a Solar PV panel glass cover [20,28]. In order to create TiO2 thin films with high surface area and unique optical and photocatalytic properties we used block copolymer assisted evaporation induced self-assembly (EISA) method. This method has been used since a long time to produce highly ordered colloidal arrays [29]. The method adopted by us makes use of high molecular weight 128 Solar Energy Materials and Solar Cells 188 (2018) 127–139 D. Adak et al. wavelength-dispersive X-ray spectrometry (WDX) or by energy-dispersive X-ray spectrometry (EDX) with a scanning electron microscope (SEM) [36]. The analytic result using this technique has been proved using a model TiO2 thin film with homogenous controlled porosity and film thickness synthesized using block copolymer template. TiO2 being wide band gap (Eg ~3.2 eV) semiconductor, it is photoactive in the UV region of solar spectrum only. In order to enhance the self-cleaning activity of TiO2 thin film and to achieve enhanced antireflection behavior, in our earlier published work we have reported photo-catalytic property of V doped TiO2-SiO2 mixed sol coatings [37]. Here we have experimented with Nitrogen (N) which is one of the most studied non-metals for extending photoactivity of TiO2 in the visible region of solar spectrum. In an exhaustive review paper by Asahi et al. N-doped TiO2 the related science and technology have been intensively reported [38]. In order to prepare N doped TiO2 several route such as magnetron sputtering, pulse laser deposition (PLD) method, filtered vacuum arc deposition (FVAD) and various type of vapor deposition technique such as metal−organic CVD (MOCVD), plasma-activated CVD (PACVD), vapor deposition, plasma enhanced CVD (PECVD) were followed [4,39–45]. In our case N doped TiO2 films were obtained by treating TiO2 thin films under nitrogen plasma generated in a plasma reactor. The nitrogen doping in the TiO2 coating is verified by the change in the band gap, as obtained from Tauc's plot and EDX elemental mapping. To ensure proper condition for nitrogen incorporation we varied RF self-bias potential, time of plasma treatment and optimized the treatment conditions by performing quite a large number of experiments. The nitrogen plasma treatment is found to enhance the photo-catalytic activity which has been verified by performing photo-degradation study of a common textile dye, Methylene Blue (MB), before and after the plasma treatment. The enhanced photo-activity has been obtained without compromising the overall optical transmission and mechanical stability of such coatings which has been confirmed by nano-hardness tests. The regular porous structure of TiO2 coatings has been verified from Field-emission scanning electron microscope (FESEM), Atomic force microscope (AFM) images and ellipsometry studies. The extreme wettability of such coating has been confirmed by water contact angle measurements. Under optimized conditions we have been able to achieve super-hydrophilicity. Other physical characterization studies have been performed with the help of X-ray reflectivity (XRR), Fourier Transform infrared spectroscopy (FTIR spectroscopy), Energy dispersive X-ray (EDX) and X-ray diffraction (XRD) analysis. To investigate mechanical hardness of the coating we studied elastic modulus and hardness using a nano-indentation technique. Thus, we have been able to develop a process for fabricating single layer structured TiO2 coating with regular pores which are found to exhibit simultaneous anti-reflecting, superhydrophilic and photoactive behavior to be used as coating on PV solar panel. Fig. 1. Mechanism and synthesis procedure of porous TiO2 thin film. thin film coatings is shown in cartoon given in Fig. 1. The triblock copolymer Pluronic F127 has been used as structure directing agent in this study. The films were prepared from solution containing TBOT/ F127/HCl/H2O/EtOH with respective molar ratio of 1:0.005:1.75:10:50. At first, 0.42 g of F127 was dissolved in solution composed of EtOH and HCl and stirred for at least 20 min prior to addition of TBOT. After the addition of TBOT, H2O was added in the solution and the entire solution was further stirred for 2 h. Finally, the solution was left for aging for at least 10 days after addition of 5 wt% acid catalyzed SiO2 sol prepared by a method reported by Zhou et al. [46]. SiO2 sol helps to stabilize the TiO2 sol extending the gel point over few months. The same sol has been used for months, for coating films by both dip coating and spin coating techniques. The glass substrates used for coating were first cleaned by mild brushing, followed by sonication for 30 min in 5% Extran® solution dissolved in DI water. After thoroughly rinsing the glass substrates in DI water they were degreased by consecutively boiling in a TCE, Acetone and IPA solution. Finally, the substrates were dried in an air-oven at 100 °C for 30 min before using them for film coating. In order to optimize coating conditions, we varied dipping speed from 60 mm/min to 300 mm/min. Right after coating the sol, it was treated in a muffle furnace for proper crystallization of TiO2 sol coated glass substrates. The coated films were first heated at 250 °C for 60 min with ramping rate of 1 °C/min in air. Such slow heating was required for complete removal of organic polymer from the surface of thin films. Finally, the films were annealed at 400 °C for 2 h. The furnace was allowed to cool down to room temperature and the films were taken out for further experiments. In order to ensure no residue was left over the coating, it was sonicated in IPA for 5 min followed by heating in the air-oven at 100 °C for 5–10 min. 2. Material and methods 2.1. Materials 2.2.2. Coating of sol on glass In order to ensure the coating parameters were well optimized the dipping and withdrawal speed of porous TiO2 were varied from 60 mm/ min to 300 mm/min. Each deposition was followed by annealing as discussed above. Finally, the transmission and water contact angle of coatings were measured. From Fig. 2(a) it can be clearly seen that maximum transmission is obtained in case of coatings deposited at the deposition and withdrawal rate of 200 mm/min. On increasing the deposition rate from 60 mm/min to 100 mm/min the optical transmission increases from nearly 92.8–94.1% and finally on increasing the speed to 200 mm/min the maximum transmission of 95.1% is observed. On further increasing the withdrawal speed (from 200 mm/min to 300 mm/min) the optical transmission is found to be 93.6%. The Tetrabutylorthotitanate (TBOT), Tetraethyl orthosilicate (TEOS) and Triblock copolymer Pluronic® F127 (HO-(CH2CH2O)106(CH2CH (CH3)O)70(CH2CH2O)106H, Molecular weight = 12,600) were purchased from Sigma-Aldrich. Methylene Blue (MB) dye was purchased from Spectrochem & Co. Ethanol (EtOH), Trichloroethylene (TCE), Isopropanol (IPA), Acetone, Extran® and Hydrochloric acid (HCl) were purchased from Merck & Co. All the chemicals were of analytical grade. 2.2. Optimization of porous TiO2 thin films 2.2.1. Preparation of sol The technique adopted for the preparation of self-patterned TiO2 129 Solar Energy Materials and Solar Cells 188 (2018) 127–139 D. Adak et al. Fig. 2. (a) Optical Transmission spectra and (b) WCA Vs. withdrawal speed of dip coater for preparing optimized porous TiO2 thin film. UV–VIS–NIR Spectrophotometer (Shimadzu, SolidSpec-3700) and Photoluminiscence (HORIBA Jobin Yvon Nanonlog). The material composition of samples was analyzed by EDX system (JSM7610F, JEOL, Japan). Ellipsometry measurements were performed using Spectroscopic Ellipsometry (VASE32, J A WOOLLAM, USA) to study refractive index and thickness of the coatings. Water contact angles (WCAs) of coatings were measured, at room temperature, with a Goniometer (290 G1, Ramehart, USA). The coating hardness was investigated using Nanoindentor (Hysitron TI 950, Bruker). The photocatalytic self-cleaning property of the coatings was evaluated by investigating their ability to decompose MB dye in the presence of light. Light source used in the photodegradation experiment was a mercury vapor lamp (58 lm/watt) that was calibrated to match the intensity of ambient solar power, in this spectral window, using a lux meter. A MB stock solution of 50 ppm was prepared and preserved in dark. The optimized coating was obtained by dip coating TiO2 sol at 200 mm/min followed by annealing. For the photodegradation study TiO2 films, before and after plasma treatment, 55 W power for 2.50 min was chosen. The films were coated with as prepared 50 ppm MB dye solution by dip coating the films at the rate of 60 mm/min. The photodegradation of the coatings were evaluated by measuring transmission spectra with UV–vis spectrophotometer (JASCO 750V) by studying the evolution of a trough due to absorption of MB dye over a period of time. maximum optical transmission of bare glass is ~ 91.8%. Also, the water contact angle of the coating prepared by varying withdrawal speed has been investigated. As can be seen from Fig. 2(b) the minimum water contact angle ~ 5° is found in case of films coated with a dipping speed of 200 mm/min. The films that are being investigated in this paper to study self-cleaning activity and anti-reflecting behavior were deposited by dip coating at the withdrawal speed of 200 mm/min. 2.2.3. Plasma treatment To enhance the photo-catalytic activity of porous TiO2 thin film we attempted to dope nitrogen by plasma assisted technique in a modified RF (13.56 MHz) sputtering system using ultra-high purified N2 (99.999%) gas. Islam et al. had experimented with the microwave assisted plasma CVD system for the N doping of TiO2 films [41]. To simplify the process of N doping in the porous TiO2 coating we made certain changes to the existing RF Magnetron sputtering unit by first removing the ceramic magnets and the substrates were kept at the powered electrode in a so called etching geometry. The developed DC self-bias is, thus, an indicator of energy of N ion/radical impacting on the film. The sample was placed on an alumina boat placed on the cathode. The chamber was evacuated until it reached a vacuum level of as low as 5 × 10−6 mbar. Once a base pressure was reached and stabilized, N2 gas was introduced into the chamber with a flow rate of 40 SCCM (standard cubic centimeter per minute) for ~ 10 min. After purging with N2 gas the throttle valve was turned to 15% open position, to maintain the chamber pressure at 4.7 × 10−2 mbar (process pressure). Finally, RF power of 55 W was applied to the cathode and N2 plasma generated. The self-bias potential was noted down for each process run. Plasma treatment duration was varied from 1.50 to 10 min. The plasma treatment for 2.50 min at 55 W (485 V) has proved to be the best optimized condition, the details of which has been discussed in Section 3.4. 3.2. Structural analysis In order to perform structural characterization and determine type of bonding present in the TiO2 thin films, FTIR spectra of the films were recorded and analyzed. Four samples of TiO2 thin film coated at different withdrawal speed viz. S1 at 60 mm/min, S2 at 100 mm/min and S3 at 200 mm/min were investigated. Based on the discussion made in the previous section (Section 2.2.2) the samples S3 were found to be best optimized and plasma treated for further investigations. The plasma treatment was carried out on a sample deposited at 200 mm/ min for 2.50 min at D.C self-bias potential 485 V (RF power 55 W). This sample is denoted as S4. A sharp peak around 1100 cm−1 can be observed in all four samples which can be attributed to Si-O-Si stretching and the small peaks around 700–950 cm−1 can be assigned to Ti-O-Si and Ti-O-Ti vibration frequencies [47,48]. The peaks centered at 3300 cm−1 are, presumably, due to O-H bonds adsorbed at the surface of the coating. In comparison to pure TiO2 coating (S1, S2 and S3) the plasma treated TiO2 coating (S4) shows an extra peak around 2352 cm−1 which can be attributed to the N atoms embedded in the TiO2 network in the form of O-Ti-N bonds [49]. Further, presence of 3. Results and discussion 3.1. Characterization studies The morphologies and surface roughness of TiO2 coatings were examined by a Field-emission scanning electron microscope (FESEM JSM7610F, JEOL, Japan) and an Atomic force microscope (5100, Agilent Technologies, USA). The structural characterization of thin films was done using X-ray reflectivity (Rigaku Ultima IV, multipurpose X-Ray Diffraction System, source Cu Kα, λ = 0.1540 nm), FTIR (Perkin Elmer 100 spectrophotometer) and Raman spectroscopy (WITec alpha 300). The optical transmission measurements were carried out using a 130 Solar Energy Materials and Solar Cells 188 (2018) 127–139 D. Adak et al. 3.3. Morphological analysis A regular porous structure can be clearly seen from FESEM micrographs in Fig. 4. It is evident from these images that the coatings consist of interconnected nanoporous structure and regular nanotextured surfaces. The higher transmission and superhydrophilicity of self-assembled TiO2 coating can be attributed to such structural features as shown in Fig. 4. From the micrographs of plasma treated TiO2 thin films, it is clearly evident that there is almost no change in the structural features of such coatings, besides enlargement of surface area due to feature broadening after plasma treatment. From the EDX images embedded in the inset of FESEM micrograph we obtain clean indication of nitrogen incorporation after plasma treatment. All the films are also found to be rich in oxygen content. The plasma treated samples indicate presence of Ti, O, and N while on the other hand no N content is found in case of untreated TiO2 coatings. The elemental mapping in the inset of EDX spectra also confirms the presence of N, O and Ti atoms in the coatings. Thus, we infer a regular porous structure by using structure directing agent (Pluronic F127) and successfully incorporated N by treating, as prepared thin films, under N plasma. The Atomic Force Microscope (AFM) images in Fig. 5 highlights the variation in surface topography of the TiO2 thin films upon addition of block copolymer and the change in morphology upon exposure to N plasma. The RMS roughness (Ra) of each sample is mentioned in the respective images. It is observed formation of nanoscale features exclusively depends on sol gel precursor that contains block copolymer and Ti precursor. The surfaces of the TiO2 thin films are found to be smooth, with low roughness values, which may be attributed to the hydrophilicity of these coatings (RMS roughness ~ 1.65 nm). It can also be seen from Fig. 5(c) that under nitrogen plasma treatment for 2.50 min at 55 W the roughness is found to reduce from 1.65 nm to 1.52 nm which maybe the reason of enhanced superhydrophilicity. The roughness increases on increasing the plasma treatment time. On the other hand, while treating the TiO2 thin film for shorter time ~ 1.50 min (Fig. 5(b)); there is a slight increase in the average surface roughness as compared to untreated thin film. Again, films treated under plasma for longer time viz. 2.50 min the average roughness decreases slightly. The decrease in roughness may presumably be responsible for very low water contact angle (~ 2.5°). Further, on increasing plasma treatment time lead to increase in surface roughness, in turn, causing increase in water contact angle ~ 12°. The nature pore distribution and approximate pore diameters of these TiO2 films can be clearly seen from the 2D AFM images in ESI (Fig. ES10). To check the repeatability of our experimental findings, RMS roughness is calculated at over 10 locations on each sample. The standard deviation of the roughness data is found to be ≤ 0.025 nm, which shows uniform morphology across the entire sample. Error bar graph depicting variation of RMS roughness with nitrogen plasma treatment time of porous TiO2 thin film has been given in Fig. ES9 (ESI). Hence, it can be said that the thin films treated for 2.50 min, under nitrogen plasma can be taken as the best optimized coating, which exhibits superhydrophilicity while at the same time maintain high optical transmission. This is further discussed in detail in the later sections. Fig. 3. FTIR spectra of untreated (S1, S2 and S3) and N plasma treated (S4) TiO2 thin films. TiO2 has been confirmed by Raman spectral analysis and the corresponding plot is given in Fig. ES3 in Electronic Supplementary information (ESI). It has been found from the peak position in Raman spectra that the values for Raman modes agree well with the literature values for anatase TiO2. The detailed analysis has been given in ESI. From the Raman data certain peaks corresponding to SiO2 are also noticed but one cannot unequivocally say that they relate to SiO2 present in the body of the film or originate from substrate material, which is glass (Fig. 3). In order to investigate surfaces of the thin TiO2 coatings we performed X-ray reflectivity analysis of these films. It is a well known technique believed to offer accurate thickness values for thin films and multilayer as well as density, surface and interface roughness. The detailed procedure of this technique and governing principle has been discussed in detail by Isao et al. [50]. The X-ray reflectivity profiles measured using our XRD set up and the corresponding fitting curve and model for TiO2 thin film has been given in the ESI (Fig. ES5, Fig. ES6). XRR data were recorded to obtain thickness, density and average roughness of the TiO2 thin films, and is summarized in Table 1. From Table 1 it is evident that as we increase the withdrawal speed while coating, thickness increases and at the same time density of TiO2 thin films is found to decrease as compared to bulk TiO2 [29]. In case of S4 which is obtained by treating the TiO2 thin films deposited at the speed of 200 mm/min, we observe a slight decrease in the thickness as well as density of thin films. The decrease in the density of TiO2 thin films may be attributed to the enhanced porosity of the films which is evident from ellipsometry studies (Section 3.5). Besides we also obtained average roughness value which closely matches with the roughness value obtained from AFM study. 3.4. Effect of plasma treatment In order to improve the performance of TiO2 thin film, it was treated under nitrogen plasma. The detailed experiment has been discussed in previous Section 2.2.3. To get best optimized result we varied the treatment conditions like treatment time and self-bias potential. As evident from both Fig. 6 and Fig. 7 the best result is obtained in the case of plasma treatment for 2.50 min, after which the transmission of TiO2 films start dropping. As mentioned in Section 2.2.2 the best optimized coating has been designated as S4. The samples are designated on the basis of treatment time value (S4_XX min) and RF power value (S4_YY Table 1 XRR data of TiO2 thin films coated under different conditions. Sample Name Thickness (nm) Density (g/ cm3) Average roughness (nm) S1 S2 S3 S4 63.0 83.3 94.7 82.0 3.10 2.97 2.80 2.70 1.08 1.56 1.93 1.30 131 Solar Energy Materials and Solar Cells 188 (2018) 127–139 D. Adak et al. Fig. 4. FESEM images of (a) untreated and plasma treated samples at 55 W for (b) 1.50 min, (c) 2.50 min and (d) 4 min. W), where XX and YY denotes the value for treatment time and power applied, respectively. Fig. 6. indicates that nitrogen plasma treatment does not affect the overall transmission of TiO2 thin films, in case of treatment time of 1.50 min and 2.50 min. On further increase of treatment time the transmission around 500–700 nm drops which may be attributed to the change in the surface property like roughness and nanostructure leading to an enhanced scattering or absorption. It can be clearly seen from Fig. 7(a) that the maximum band shift could be obtained in case of treatment for 2.50 min. The water contact angle value of TiO2 films treated under N plasma, under different conditions, has been shown in Fig. 7(d). From Fig. 7(d) it can be also noted that maximum hydrophilicity (WCA ~ 2.5°) was achieved in the case of treatment time of 2.50 min. The bias potential for plasma treatment, under such conditions, was found to be around 485–488 V. The band gap of porous TiO2 films has been found to shift from 0.1 to 0.3 eV towards the visible part of the solar spectrum, which may be attributed to the N incorporation in the TiO2 lattice. Further, change in band-gap by changing process time has been shown in Fig. 7(b). It can be clearly seen from Fig. 7(b) that the maximum change in the bandgap i.e. 0.3 eV is observed in the case of plasma treatment time of 2.50 min, at 55 W applied power, corresponding to a self-bias voltage of 485 V. The details of the findings have been summarized in Table 2, from where it can been found that the band-gap change for other process times is very small, as compared to that of process time ~ 2.50 min. To validate the role of self-bias potential the plasma power is varied by keeping the time of treatment constant at 2.50 min. In this case, it can be observed from Table 3 that the maximum shift in band gap energy has been obtained for plasma power = 55 W and corresponding D.C Self bias potential = 485 V. The detailed plasma treatment conditions and the corresponding plasma power values have been further summarized in Table 3. Thus, it can be said that maximum shift in the band energy has been achieved, without affecting overall transmission for plasma treatment at D.C. bias potential of 485 V for 2.50 min. To, further investigate effect of N plasma treatment on porous TiO2 thin films; we carried out PL spectral analysis. It can be seen from Fig. ES4, given in the ESI,that N plasma treated TiO2 samples show a higher PL intensity than untreated TiO2. This may presumably be due to an increase in the total amount of photogenerated charge carriers and surface defects that can contribute to the PL emissions [51]. Also, the PL spectra of N plasma treated samples are found to be little red shifted, indicating change in the band gap (reduction in optical band-gap) which can be attributed to the N doping through plasma treatment [52]. Further, detailed discussion can be found in the ESI section. 3.5. Optical study using spectroscopic ellipsometry The refractive index (η) and extinction coefficient (k) were obtained by spectroscopic ellipsometry (SE) measurements. A three-phase model of air/porous TiO2 film/Si (100) substrate has been used in the simultaneous fitting of the measured parameters Δ and Ψ of SE. Fitting is based on the simple Cauchy model. Since the refractive index of glass and that of porous TiO2 are almost similar, coatings obtained were found to be very transparent and signals obtained were buried in noise. Therefore, we used polished Si wafer, instead of glass substrates, for our further investigations. The thickness values of the thin films obtained from SE study varies slightly from those obtained from XRR study due, presumably, to the different substrates used in these experiments. Four samples, namely untreated porous TiO2 (S4), porous TiO2 treated under N2 plasma for 1.50 min (S4_1.5 min), 2.50 min (S4_2.5 min) and 4.0 min (S4_4 min) were characterized using this technique. It is clearly evident from Fig. 8(a) that refractive index of porous TiO2 films are substantially lower than that of bulk TiO2 films, prepared by sol gel method in previously reported work [53,54]. A further decrease in the refractive index was found on treating porous TiO2 thin films under N plasma under different conditions as reported in the previous section. For example S2, the refractive index n = 1.27 is lower than that of untreated porous TiO2 thin film with n = 1.35 at 550 nm. Such lower value of refractive index is obviously due to the porous structure obtained by using a block copolymer as a structure directing agent while 132 Solar Energy Materials and Solar Cells 188 (2018) 127–139 D. Adak et al. Fig. 5. AFM images of (a) untreated and plasma treated samples for (b) 1.50 min, (c) 2.50 min and (d) 4 min, with RMS roughness (Ra) in the inset of each image. increase of the treatment time causes increase in the refractive index, thereby, resulting in a lower transmission as compared to untreated porous TiO2 thin films. Plasma effects on coated porous films indeed involve several processes and defy straight forward interpretation. However, it can be said that the gross effect of the plasma treatment appears to be first removal of certain absorbed/adsorbed impurities (residual polymer species, etc.). Such absorbed species appear to block the pores leading to a effective refractive index which is higher. Initial plasma processing (for 1–2.5 mts) clears the pores and the refractive index is lowered because of resultant comparatively higher porosity. It is clearly evident from the 2D AFM images in the ESI (Fig. ES10), as the plasma treatment time increases pore diameter are found to increase and after prolonged treatment (say, about 4 min in this case) there is a drastic increase in the pore diameter. This means per unit area there are now lesser number of pores. This enhances the density of the films leading to an increase in the refractive index. synthesizing the films. To evaluate the porosity in the respective TiO2 thin films we have used the following equation based on Bruggeman effective medium approximation method [55]. n2−1 Porosity (%) = ⎜⎛1− 2 ⎟⎞ × 100 n d −1 ⎠ ⎝ (1) Where, nd is the refractive index of pore-free TiO2 (n = 2.25 at 550 nm), and n is the refractive index of the porous TiO2 thin films. To validate our finding we refer to an already reported TiO2/SiO2 bilayer thin film system prepared using pluronic F127 block copolymer [35]. Here, the 80% voids have been found in case of SiO2 underlayer and the top TiO2 layer comprises of 54% voids. The refractive index data for this bilayer system, which is in the range of 1.14–1.38 for the bottom layer and 1.52–2.55 for the top layer closely matches with calculated voids and at the same time is in good agreement with that of reported optical calculations. In order to calculate porosity in our as prepared thin films using block copolymer we used refractive index data (nd) of TiO2 thin films prepared by sol gel coating without using any block copolymer template [56]. In our case as given in Table 4 and Fig. 8, porosity of TiO2 thin films prepared using block copolymer is in the range of 75–80% which has been further enhanced, up to 85% by nitrogen plasma treatment. The void fraction/ porosity variation with sample has been depicted in the inset of Fig. 8(b). The enhanced porosity has also been reflected in the corresponding transmission data (Fig. 6). We have found reduction in refractive index on plasma treatment as compared to untreated samples up to 2.50 min. Further 3.6. Self-cleaning activity by photo-degradation of organic pollutants Superhydrophilic surfaces often exhibit self-cleaning behavior. The dirt rides piggyback and washes away with water as water easily seeps in between the superhydrophilic coating and attached dirt particles. Thus, such self-cleaning coating provides hydrophilic sheathing to the material on which it is applied [37]. We have also depicted pictorially in Fig. 9 that how the self-cleaning behavior is realized with 133 Solar Energy Materials and Solar Cells 188 (2018) 127–139 D. Adak et al. Fig. 6. Transmission spectra for N2 plasma treated sample by varying time of treatment - (a) 1.50 min, (b) 2.50 min, (c) 4 min and (d) 10 min. electrons from the valence band, as the TiO2 band gap is around 3.0 eV. After N doping, the electron can be excited from the N impurity levels to the conduction band. The N impurity level is directly responsible for the origin of visible light induced photocatalysis. A process of visible light photocatalytic oxidation of MB dye is described in Fig. 9. The detailed mechanism of the same has been reported in many earlier published literatures [38,45,61]. To evaluate the visible light photocatalytic activity of the as prepared samples, photodegradation of MB dye was performed. The results of the photocatalytic activities are shown in Fig. 10. In our previous work we have found that in case of TiO2 thin film prepared without using any block copolymer the photodegradation of MB dye takes place in more than 3–4 h [37]. In order to investigate effect of N plasma on TiO2 thin film, the photocatalytic study has been made on the best optimized sample deposited at 200 mm/min (S4_0 min) and plasma treated sample (S4_2.5 min) for 2.50 min. Photocatalytic degradation of samples deposited under different conditions has been carried out and the corresponding results have been given in Fig. ES12 and Fig. ES13 in the ESI. It can be seen from Figs. 10(a) and 10(b) that N doped TiO2 sample is superior to untreated sample as the degradation time in the case of plasma treated film is 70 min, while untreated TiO2 film is 100 min. The dye degradation has been observed by monitoring disappearance of trough around 663 nm which is a characteristic of MB dye and increase in transmission with time. This observation is further confirmed by the rate constant of MB dye degradation which follows a first order kinetics. Untreated samples exhibit lower photocatalytic activity (k = 0.00108 min−1) under visible light irradiation due to its weak superhydrophilic surfaces, utilizing the photocatalytic properties of TiO2 [57]. The self-cleaning effect of photocatalytic TiO2 surfaces is based on the absorption of light in the ultraviolet spectral range, λ < 375 nm, exciting charge carriers to the conduction band in a metal oxide semiconductor with wide band gap energy (Egap~3.2 eV). It has been reported that band structure engineering has been implemented by narrowing the band gap of TiO2 based material, mainly by doping and allowing (detailed discussion provided in the introduction) which significantly extends the light absorption also in the visible light range thereby enhancing absorption in a larger part of the solar spectrum [58]. Liu et al. have reported band shift up to 2.99–2.75 eV by treating TiO2 thin films prepared using ALD technique by thermal treatment in NH3 flow at 600 °C for 1–4 h [59]. Also, there is a report of achieving band gap of 2.95 eV by synthesizing TiO2 powder via hydrothermal method at 150 °C in the presence of urea and guanidine, which acts as a nitrogen source [60]. In our case, nitrogen doping has been achieved at room temperature using nitrogen gas by following a plasma technique as discussed in the previous section. With the help of nitrogen plasma treatment we have been able to achieve the band gap of 2.94 eV, thereby, extending the photocatalytic activity up to 421.7 nm. On illuminating N doped TiO2 thin film with mercury vapor lamp some of the photo-excited charge carriers could obviously react with the atmospheric oxygen and water molecule to form hydroxyl radical (OH•) and superoxide radical anion (O2-•) radicals, which help to degrade nearby organic molecules. This “cold combustion” mechanism enables self-cleaning through the conversion of organic pollutants to carbon dioxide, water, and mineral acids. When there is no N doping by plasma treatment the energy of the visible light is not sufficient to excite 134 Solar Energy Materials and Solar Cells 188 (2018) 127–139 D. Adak et al. Fig. 7. (a) Tauc's plot for N2 plasma treated for 1.50 min (b) 2.50 min, (c) 4 min and (d) corresponding WCA for TiO2 coating treated for different time. visible light absorption owing to high band-gap energy. However, the plasma treated samples exhibit enhanced photocatalytic activity (k = 0.00134 min−1) as compared to the untreated ones. This has been further reflected in the degradation percentage and recovery percentage of MB dye as shown in Fig. 10(c) and Fig. 10(d). The MB degradation and percentage recovery of TiO2 thin films treated under nitrogen plasma for different time interval has been further depicted in Fig. ES12 and ES13, respectively, in the ESI. Hence, we infer that on plasma treatment of the TiO2 coatings show enhanced self-cleaning activity due to its pronounced photocatalytic activity. It has been found that as we increase the nitrogen plasma treatment time there occur increase in photodegradation rate as compared untreated TiO2 thin film. A detailed comparison table (TS2) has been put up in the ESI detailing photodegradation ability of several related systems with that of present coatings. Table 3 Nitrogen plasma treatment conditions relative to plasma power and corresponding change the band gap energy. Sample Name Applied RF Power (W) Time (min) D.C. Self Bias Voltage (V) Eg (eV) Before N2 Plasma Treatment Eg (eV) After N2 Plasma Treatment S4_45W S4_55W S4_65W S4_100W 45 55 65 100 2.50 2.50 2.50 2.50 435 485 523 652 2.97 2.97 2.97 2.98 2.95 2.94 2.96 2.97 many labs and industries, for testing mechanical hardness is the pencil hardness test [62]. It has been found that in this case samples have passed 3 H pencil test. This satisfies the initial requirement for undertaking further more rigorous testing. The pencil hardness value of 5 H has been obtained in case of double layer coatings or coating composed of mixed composition consisting SiO2 and TiO2 sol [63,64]. While, on the other hand, the hardness value of 3 H can be considered good enough for a single layer coating as it is in our case [28]. The intrinsic 3.7. Mechanical hardness of TiO2 thin film The mechanical properties of thin films on glass substrates were tested first by pencil hardness test qualitatively, and then more rigorously using a nanoindentor. A practical approach adopted often in Table 2 Nitrogen plasma treatment conditions relative to treatment time and corresponding change the band gap energy. Sample Name Power Applied RF (W) Plasma Treatment Time (Min) D.C Self Bias Voltage (V) Eg (eV) Before N2 Plasma Treatment Eg (eV) After N2 Plasma Treatment WCA (θ°) S4_0 min S4_1.5 min S4_2.5 min S4_4 min – 55 55 55 0 1.50 2.50 4.00 – 485 485 485 2.97 2.97 2.97 2.99 2.97 2.96 2.94 2.98 4.9 6.0 2.5 10 135 Solar Energy Materials and Solar Cells 188 (2018) 127–139 D. Adak et al. Fig. 8. The plot of data obtained from spectroscopic ellipsometry; (a) refractive index of four porous TiO2 samples coated on polished Si wafer and (b) respective porosity for refractive index at 550 nm, wavelength. hardness values. Fig. 11 shows the load−depth curves of the coatings tested under the same loading/unloading conditions. The elastic modulus was calculated by the change of applied force with the indentation depth. It can be seen that for a given applied load > 40 μN there is an increase in the indentation depths for untreated TiO2 thin films compared to nitrogen plasma treated TiO2 to untreated TiO2 thin films which indicate enhanced hardness in case of plasma treated samples. The hardness and reduced elastic modulus are given by Eqs. (2) and (3) [66]. Table 4 Optical parameters from spectroscopic ellipsometry study of porous TiO2 thin films, before and after Nitrogen plasma treatment. Sample Name Refractive Indices at 550 nm Extinction Coefficient At 550 nm Porosity (%) Thickness (nm) Determined by SE S4_0 min S4_1.5 min S4_2.5 min S4_4 min 1.35 1.27 1.31 1.45 0.041 0.059 0.054 0.016 79.7 84.9 82.3 72.8 53.2 64.4 59.3 45.1 H= Pmax A (hc ) (2) Where H is hardness, Pmax is the maximum force and A(hc) is the contact area. hardness and Young's modulus of the coatings were measured using a nanoindenter. Although it is often suggested that a film thicker than 500 nm should be used to avoid the effect of the substrates used, Berasategui et al. showed that the nano-indentation technique could be as well applied to coatings less than 100 nm as well [65]. Therefore, we understand the results obtained in this case could be taken as actual π ⎞ Er = ⎛⎜ ⎟ × S 2 A (hc ) ⎠ ⎝ (3) Where Er is elastic modulus, A(hc) is the contact area and S is stiffness. Fig. 9. Band structure of nitrogen doped TiO2 and cartoon of the visible light photocatalytic process. 136 Solar Energy Materials and Solar Cells 188 (2018) 127–139 D. Adak et al. Fig. 10. Photocatalytic degradation with time for (a) untreated TiO2 (S4_0min), (b) plasma treated TiO2 thin films (S4_2.5 min), (c) photodegradation and (d) recovery percentage of untreated and plasma treated TiO2 vs. time of irradiation. in published literatures. As reported by Zou et. al. the hardness parameter for double layer SiO2 coatings with dense acid catalyzed SiO2 bottom layer and porous F127-SiO2 layer are 2.3 GPa and 48 GPa, respectively [46]. In a similar work reported by Miao et al. the hardness value of bilayer coating consisting of bottom F127-SiO2 layer and top F127-TiO2 layer is 5.98 GPa [35]. In this case the bottom SiO2 layer comprises of nearly 80% voids clearly indicating that the porous skeleton of such films doesn’t deteriorate the mechanical properties of such thin films. Thus, it can be said that the porous TiO2 samples exhibit sufficient mechanical hardness to be used on solar glass cover and the mechanical strength is further enhanced on nitrogen plasma treatment. In addition to mechanical strength, to check the durability of these coatings we have kept the thin films in an open air for over one month and measured the transmission once after every seven days. We found slight decrease in the transmission during each measurement which could be completely regained on keeping the samples in ultrasound sanitation for 2–3 min. Hence, such coatings are useful for outdoor application and aid to maintain the transparency of solar panels. Further, to check adhesion strength of the coatings, the representative thin film coated on glass substrates have been kept submersed in 5% saline water and boiled for over 60 min. There is no peeling/delamination or change observed in the optical transmission or contact angel values, before and after performing such experiments, which clearly indicates that the coating can sustain outdoor environment. This adhesion test has been mentioned in a report published online for coating on solar glass covers by a reputed manufacturer from China [16] Similarly, the samples are found to be UV stable as we found no change in the transmission of coating on exposure to 24 W 230 nm UV Fig. 11. Load−depth curves of porous TiO2 and nitrogen doped TiO2 coatings on glass substrates. To cross check the hardness parameter using nanoindentor the measurements were taken across the samples on 44 points. The hardness and elastic modulus values for porous TiO2 thin film falls in the range 0.86–1.17 GPa and 38.8–63.4 GPa, respectively. On treating these films under nitrogen plasma we found a some improvement in the mechanical properties of thin films, resulting in hardness and elastic modulus values in the range of 0.92–2.02 GPa and 30.4–47.7 GPa, respectively. To cross check these values we compare with those available 137 Solar Energy Materials and Solar Cells 188 (2018) 127–139 D. Adak et al. lamps overnight. It appears such single component structured Titania based coating can be further improved for commercial application by more cost-effective techniques capable of volume production over large areas. Hence, such coatings are useful for outdoor application and aid to maintain the transparency of solar panels. It is expected that further tailoring and optimization of various process parameter will lead to a more robust coating for a range of applications involving self cleaning. The film strain and stress generated in untemplated TiO2 films has been calculated using Williamson-Hall method from GIXRD of thicker TiO2 films. The stress generated in the tetragonal anatase TiO2 lattice ~ 0.156 GPa. This values matches well with that of value reported in the literature relating to thin films prepared by sol-gel method [67,68]. Again, this value is significantly lower than that of block copolymer template TiO2 thin films. Detailed explanation has been given in ESI. References [1] H. Zhong, Y. Hu, Y. Wang, H. Yang, TiO2/silane coupling agent composed of two layers structure: a super-hydrophilic self-cleaning coating applied in PV panels, Appl. Energy 204 (2017) 932–938. [2] M.R. Maghami, H. Hizam, C. Gomes, M.A. Radzi, M.I. Rezadad, S. Hajighorbani, Power loss due to soiling on solar panel: a review, Renew. Sustain. Energy Rev. 59 (2016) 1307–1316. [3] H.K. Elminir, A.E. Ghitas, R.H. Hamid, F. El-Hussainy, M.M. Beheary, K.M. AbdelMoneim, Effect of dust on the transparent cover of solar collectors, Energy Convers. Manag. 47 (2006) 3192–3203. [4] H. Yan, W. Yuanhao, Y. Hongxing, TEOS/silane coupling agent composed double layers structure: a novel super-hydrophilic coating with controllable water contact angle value, Appl. Energy 185 (2017) 2209–2216. [5] I.P. Parkin, R.G. Palgrave, Self-cleaning coatings, J. Mater. Chem. 15 (2005) 1689–1695. [6] S.R. Hunter, D.B. Smith, G. Polizos, D.A. Schaeffer, D.F. Lee, P.G. Datskos, Low cost anti-soiling coatings for CSP collector mirrors and heliostats, in: High and Low Concentrator Systems for Solar Energy Applications IX, International Society for Optics and Photonics, 2014, p. 91750J. [7] W. Barthlott, C. Neinhuis, Purity of the sacred lotus, or escape from contamination in biological surfaces, Planta 202 (1997) 1–8. [8] R.A. Fleming, M. Zou, Silica nanoparticle-based films on titanium substrates with long-term superhydrophilic and superhydrophobic stability, Appl. Surf. Sci. 280 (2013) 820–827. [9] K. Midtdal, B.P. Jelle, Self-cleaning glazing products: a state-of-the-art review and future research pathways, Sol. Energy Mater. Sol. Cells 109 (2013) 126–141. [10] D. Lee, M.F. Rubner, R.E. Cohen, All-Nanoparticle Thin-Film Coatings, Nano Lett. 6 (2006) 2305–2312. [11] V. Zorba, X. Chen, S.S. Mao, Superhydrophilic TiO2 surface without photocatalytic activation, Appl. Phys. Lett. 96 (2010) 093702. [12] M. Takeuchi, K. Sakamoto, G. Martra, S. Coluccia, M. Anpo, Mechanism of Photoinduced Superhydrophilicity on the TiO2 Photocatalyst Surface, J. Phys. Chem. B 109 (2005) 15422–15428. [13] J. Son, S. Kundu, L.K. Verma, M. Sakhuja, A.J. Danner, C.S. Bhatia, H. Yang, A practical superhydrophilic self cleaning and antireflective surface for outdoor photovoltaic applications, Sol. Energy Mater. Sol. Cells 98 (2012) 46–51. [14] J. Yu, S. Chary, S. Das, J. Tamelier, N.S. Pesika, K.L. Turner, J.N. Israelachvili, Gecko‐inspired dry adhesive for robotic applications, Adv. Funct. Mater. 21 (2011) 3010–3018. [15] S. Prabhu, L. Cindrella, O. Joong Kwon, K. Mohanraju, Superhydrophilic and selfcleaning rGO-TiO2 composite coatings for indoor and outdoor photovoltaic applications, Sol. Energy Mater. Sol. Cells 169 (2017) 304–312. [16] M.A.M.Ld Jesus, J.Td.S. Neto, G. Timò, P.R.P. Paiva, M.S.S. Dantas, Ad.M. Ferreira, Superhydrophilic self-cleaning surfaces based on TiO2 and TiO2/SiO2 composite films for photovoltaic module cover glass, Appl. Adhes. Sci. 3 (2015) 5. [17] J.M. Gordon, Solar Energy: the State of the Art: ISES Position Papers, James & James, London, 2001. [18] F. Li, Q. Li, H. Kim, Spray deposition of electrospun TiO2 nanoparticles with selfcleaning and transparent properties onto glass, Appl. Surf. Sci. 276 (2013) 390–396. [19] J.-J. Park, D.-Y. Kim, J.-G. Lee, D. Kim, J.-H. Oh, T.-Y. Seong, M.F.A.M. van Hest, S.S. Yoon, Superhydrophilic transparent titania films by supersonic aerosol deposition, J. Am. Ceram. Soc. 96 (2013) 1596–1601. [20] Z. Liu, X. Zhang, T. Murakami, A. Fujishima, Sol–gel SiO2/TiO2 bilayer films with self-cleaning and antireflection properties, Sol. Energy Mater. Sol. Cells 92 (2008) 1434–1438. [21] Y. Tsuge, J. Kim, Y. Sone, O. Kuwaki, S. Shiratori, Fabrication of transparent TiO2 film with high adhesion by using self-assembly methods: application to super-hydrophilic film, Thin Solid Films 516 (2008) 2463–2468. [22] M. Rahman, F. Tajabadi, L. Shooshtari, N. Taghavinia , Nanoparticulate hollow TiO2 fibers as light scatterers in dye-sensitized solar cells: layer-by-layer self-assembly parameters and mechanism, ChemPhysChem 12 (2011) 966–973. [23] Y. Lu, S. Sathasivam, J. Song, C.R. Crick, C.J. Carmalt, I.P. Parkin, Robust selfcleaning surfaces that function when exposed to either air or oil, Science 347 (2015) 1132–1135. [24] H.E. Çamurlu, Ö. Kesmez, E. Burunkaya, N. Kiraz, Z. Yeşil, M. Asiltürk, E. Arpaç, Sol-gel thin films with anti-reflective and self-cleaning properties, Chem. Pap. 66 (2012) 461–471. [25] C.J. Brinker, G.W. Scherer, CHAPTER 14 - Applications, in: Sol-Gel Science, Academic Press, San Diego, 1990, pp. 838–880. [26] K. von Niessen, M. Gindrat, Plasma Spray-PVD, A new thermal spray process to deposit out of the vapor phase, J. Therm. Spray. Technol. 20 (2011) 736–743. [27] P. Mazumder, Y. Jiang, D. Baker, A. Carrilero, D. Tulli, D. Infante, A.T. Hunt, V. Pruneri, Superomniphobic, transparent, and antireflection surfaces based on hierarchical nanostructures, Nano Lett. 14 (2014) 4677–4681. [28] Ö. Kesmez, H. Erdem Çamurlu, E. Burunkaya, E. Arpaç, Sol–gel preparation and characterization of anti-reflective and self-cleaning SiO2–TiO2 double-layer nanometric films, Sol. Energy Mater. Sol. Cells 93 (2009) 1833–1839. [29] P. Jiang, J.F. Bertone, K.S. Hwang, V.L. Colvin, Single-crystal colloidal multilayers of controlled thickness, Chem. Mater. 11 (1999) 2132–2140. [30] H. Mcleod, Thin Film Optical Filters, Institute of Physics Publishing, Arizona, 2001. [31] S. Guldin, P. Kohn, M. Stefik, J. Song, G. Divitini, F. Ecarla, C. Ducati, U. Wiesner, U. Steiner, Self-cleaning antireflective optical coatings, Nano Lett. 13 (2013) 5329–5335. [32] X.-T. Zhang, O. Sato, M. Taguchi, Y. Einaga, T. Murakami, A. Fujishima, Self- 4. Conclusion A novel technique to grow porous TiO2 films, adopting a non-lithographic technique in combination with gaseous plasma treatment, involving an inexpensive block copolymer Pluronic F127, has been developed. By optimizing carefully various process steps followed by detailed characterization studies a set of following desirable properties of the coating so developed were obtained. Coatings are highly transparent with optical transmission as high as 95%. Coatings are also superhydrophilic with water contact angle < 5°. Finally, coatings are found to be mechanically robust with micro/nano hardness in the range of 0.95–2.02 GPa. Adhesion strength of the coatings has been checked by keeping the witness coupons submersed in 5% saline water and boiling for over 60 min, without any sign of delamination, or loss of high optical transmission and superhydrophilicity behavior. It is to be noted unlike many other similar works reported in the literature, it is a single layer structured titania coating. Further, nitrogen plasma treatment helped to enhance photocatalytic activity, also in some part of the visible region of the solar spectrum, this would enable degradation of organic pollutants settling over the surface of coatings. These coatings, therefore, meet the requirements of solar PV panel glass covers as also for other automobile and architectural applications. We believe such single component structured Titania based coating can be further improved for eventual commercial application by similar cost-effective techniques capable of volume production over large areas. Acknowledgements We would like to acknowledge Department of Science & Technology, Govt of India for financial support. We would also like to acknowledge Director, IIEST, Shibpur for allowing use of other institute facilities and encouragement for undertaking this work. The authors thank Dr. Abhimanyu Singh Rana, Assistant Professor at BML Munjal University for helping us with PL spectroscopy study and carrying out measurements required for Raman analysis. We are also thankful to Dr. Mallar Ray, Assistant Professor, Ms. Susmita Biswas, Research Fellow at IIEST- Shibpur for assisting in PL measurement and helping us with the analysis of the same and Mr. Sukanta Bose, Research Fellow at IIESTShibpur, for helping us in the calculation of film stress for untemplated TiO2 coatings. Appendix A. Supplementary material Supplementary data associated with this article can be found in the online version at doi:10.1016/j.solmat.2018.08.011. 138 Solar Energy Materials and Solar Cells 188 (2018) 127–139 D. Adak et al. [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] V.J. Babu, R.P. Rao, A.S. Nair, S. Ramakrishna, Nitrogen-doped rice grain-shaped titanium dioxide nanostructures by electrospinning: frequency and temperature dependent conductivity, J. Appl. Phys. 110 (2011) 064327. [50] I.K. Boquan Li, Structural characterization of thin films by X-ray reflectivity, Rigaku Journal 16 (1999). [51] X. Li, Z. Guo, T. He, The doping mechanism of Cr into TiO2 and its influence on the photocatalytic performance, Phys. Chem. Chem. Phys. 15 (2013) 20037–20045. [52] W. Naffouti, A. Jrad, T. Ben Nasr, S. Ammar, N. Turki-Kamoun, Structural, morphological and optical properties of TiO2:Mn thin films prepared by spray pyrolysis technique, J. Mater. Sci.: Mater. Electron. 27 (2016) 4622–4630. [53] M. Mosaddeq-ur-Rahman, G. Yu, K.M. Krishna, T. Soga, J. Watanabe, T. Jimbo, M. Umeno, Determination of optical constants of solgel-derived inhomogeneous TiO2 thin films by spectroscopic ellipsometry and transmission spectroscopy, Appl. Opt. 37 (1998) 691–697. [54] H.-Q. Jiang, Q. Wei, Q.-X. Cao, X. Yao, Spectroscopic ellipsometry characterization of TiO2 thin films prepared by the sol–gel method, Ceram. Int. 34 (2008) 1039–1042. [55] B.E. Yoldas, D.P. Partlow, Formation of broad band antireflective coatings on fused silica for high power laser applications, Thin Solid Films 129 (1985) 1–14. [56] A. Bittner, R. Jahn, P. Löbmann, TiO2 thin films on soda-lime and borosilicate glass prepared by sol–gel processing: influence of the substrates, J. Sol-Gel Sci. Technol. 58 (2011) 400–406. [57] M.C. Ferrara, L. Pilloni, S. Mazzarelli, L. Tapfer, Hydrophilic and optical properties of nanostructured titania prepared by sol–gel dip coating, J. Phys. D: Appl. Phys. 43 (2010) 095301. [58] P. Adimali, W. Sibo, G. Pu-Xian, Band structure engineering strategies of metal oxide semiconductor nanowires and related nanostructures: a review, Semicond. Sci. Technol. 32 (2017) 073001. [59] K.-I. Liu, C.-Y. Su, T.-P. Perng, Highly porous N-doped TiO2 hollow fibers with internal three-dimensional interconnected nanotubes for photocatalytic hydrogen production, RSC Adv. 5 (2015) 88367–88374. [60] J. Xu, F. Wang, W. Liu, W. Cao, Nanocrystalline N-doped powders: mild hydrothermal synthesis and photocatalytic degradation of phenol under visible light irradiation, Int. J. Photo. 2013 (2013) 7. [61] Y. Cong, J. Zhang, F. Chen, M. Anpo, Synthesis and characterization of nitrogendoped TiO2 nanophotocatalyst with high visible light activity, J. Phys. Chem. C 111 (2007) 6976–6982. [62] J. Moghal, J. Kobler, J. Sauer, J. Best, M. Gardener, A.A.R. Watt, G. Wakefield, High-performance, single-layer antireflective optical coatings comprising mesoporous silica nanoparticles, ACS Appl. Mater. Interfaces 4 (2012) 854–859. [63] X. Wang, J. Shen, Sol–gel derived durable antireflective coating for solar glass, J. Sol-Gel Sci. Technol. 53 (2010) 322–327. [64] R.V. Lakshmi, T. Bharathidasan, B.J. Basu, Superhydrophobic sol–gel nanocomposite coatings with enhanced hardness, Appl. Surf. Sci. 257 (2011) 10421–10426. [65] E. G-Berasategui, S.J. Bull, T.F. Page, Mechanical modelling of multilayer optical coatings, Thin Solid Films 447–448 (2004) 26–32. [66] R. Saha, W.D. Nix, Effects of the substrate on the determination of thin film mechanical properties by nanoindentation, Acta Mater. 50 (2002) 23–38. [67] M. Manzoor, A. Rafiq, M. Ikram, M. Nafees, S. Ali, Structural, optical, and magnetic study of Ni-doped TiO2 nanoparticles synthesized by sol–gel method, International, Nano Lett. 8 (2018) 1–8. [68] B. Bharti, P.B. Barman, R. Kumar, XRD analysis of undoped and Fe doped TiO2 nanoparticles by Williamson Hall method, in: AIP Conference Proceedings, vol. 1675, 2015, p. 030025. cleaning particle coating with antireflection properties, Chem. Mater. 17 (2005) 696–700. X. Zhang, A. Fujishima, M. Jin, A.V. Emeline, T. Murakami, Double-layered TiO2−SiO2 nanostructured films with self-cleaning and antireflective properties, J. Phys. Chem. B 110 (2006) 25142–25148. M. Faustini, L. Nicole, C. Boissière, P. Innocenzi, C. Sanchez, D. Grosso, Hydrophobic, antireflective, self-cleaning, and antifogging sol−gel coatings: an example of multifunctional nanostructured materials for photovoltaic cells, Chem. Mater. 22 (2010) 4406–4413. L. Miao, L.F. Su, S. Tanemura, C.A.J. Fisher, L.L. Zhao, Q. Liang, G. Xu, Cost-effective nanoporous SiO2–TiO2 coatings on glass substrates with antireflective and self-cleaning properties, Appl. Energy 112 (2013) 1198–1205. E. Ortel, A. Hertwig, D. Berger, P. Esposito, A.M. Rossi, R. Kraehnert, V.D. Hodoroaba, New approach on quantification of porosity of thin films via electron-excited x-ray spectra, Anal. Chem. 88 (2016) 7083–7090. D. Adak, S. Ghosh, P. Chakrabarty, A. Mondal, H. Saha, R. Mukherjee, R. Bhattacharyya, Self-cleaning V-TiO2:SiO2 thin-film coatings with enhanced transmission for solar glass cover and related applications, Sol. Energy 155 (2017) 410–418. R. Asahi, T. Morikawa, H. Irie, T. Ohwaki, Nitrogen-doped titanium dioxide as visible-light-sensitive photocatalyst: designs, developments, and prospects, Chem. Rev. 114 (2014) 9824–9852. A. Borrás, C. López, V. Rico, F. Gracia, A.R. González-Elipe, E. Richter, G. Battiston, R. Gerbasi, N. McSporran, G. Sauthier, E. György, A. Figueras, Effect of visible and UV illumination on the water contact angle of TiO2 thin films with incorporated nitrogen, J. Phys. Chem. C 111 (2007) 1801–1808. L.K. Randeniya, A. Bendavid, P.J. Martin, E.W. Preston, Photoelectrochemical and structural properties of TiO2 and N-Doped TiO2 thin films synthesized using pulsed direct current plasma-activated chemical vapor deposition, J. Phys. Chem. C 111 (2007) 18334–18340. C.W. Dunnill, I.P. Parkin, N-doped titania thin films prepared by atmospheric pressure CVD using t-Butylamine as the nitrogen source: enhanced photocatalytic activity under visible light, Chem. Vap. Depos. 15 (2009) 171–174. M.Ma.T. Watanabe, Visible light photocatalysis of nitrogen-doped titanium oxide films prepared by plasma-enhanced chemical vapor deposition, J. Electrochem. Soc. 153 (2006) 186–189. L.B.E. Goldenberg, I. Zucker, R. Avni, R.L. Boxman, The effect of nitrogen partial pressure and substrate temperature on the characteristics of filtered vacuum arc deposited N:TiO2, thin films, in: Proceedings of the 13th International Conference on Plasma Surface Engineering, Garmisch-Partenkirchen (Germany), 2012. R. Asahi, T. Morikawa, T. Ohwaki, K. Aoki, Y. Taga, Visible-light photocatalysis in nitrogen-doped titanium oxides, Science 293 (2001) 269. Y. Suda, H. Kawasaki, T. Ueda, T. Ohshima, Preparation of nitrogen-doped titanium oxide thin film using a PLD method as parameters of target material and nitrogen concentration ratio in nitrogen/oxygen gas mixture, Thin Solid Films 475 (2005) 337–341. L. Zou, X. Li, Q. Zhang, J. Shen, An abrasion-resistant and broadband antireflective silica coating by block copolymer assisted sol–gel method, Langmuir 30 (2014) 10481–10486. K. Panwar, M. Jassal, A.K. Agrawal, TiO2-SiO2 Janus particles with highly enhanced photocatalytic activity, RSC Adv. 6 (2016) 92754–92764. A.N. Murashkevich, A.S. Lavitskaya, T.I. Barannikova, I.M. Zharskii, Infrared absorption spectra and structure of TiO2-SiO2 composites, J. Appl. Spectrosc. 75 (2008) 730. 139