Solar Energy Materials & Solar Cells 152 (2016) 125–132
Contents lists available at ScienceDirect
Solar Energy Materials & Solar Cells journal homepage: www.elsevier.com/locate/solmat
A study on the free radical generation and photocatalytic yield in extended surfaces of visible light active TiO2 compounds V. Raja, L. Shiamala, K. Alamelu, B.M. Jaffar Ali n Bioenergy and Biophotonics Laboratory, Centre for Green Energy Technology, Pondicherry University, Kalapet, Puducherry 605014, India
art ic l e i nf o
a b s t r a c t
Article history: Received 8 August 2015 Received in revised form 22 February 2016 Accepted 8 March 2016
Following dip coating procedure, we have fabricated extended photocatalytic surface (40 mm 25 mm) of TiO2, nitrogen doped TiO2 (N–TiO2), porous TiO2 (porous-TiO2), nitrogen doped porous TiO2 (porousN–TiO2) materials over glass slides. A red shift was observed giving a bandgap of 2.56 eV for nitrogen doped TiO2. Raman Scattering shows predominant anatase phase in all the samples along with rutile phase. Using di-chloro fluorescein (DCFH) as a fluorescence reporter molecule, we have characterized OH radical generation. Results show a 5.69 10 8 ppm/cm2/s of OH radicals are generated in porous-N–TiO2 coated surface compared to 3.69 10 8 ppm/cm2/s in TiO2 which confer a 1.54 increase in the utilization of visible light ability to the catalyst. Photocatalytic ability quantified with Methylene Blue degradation gave 1.48 fold increase under visible light exposure for porous-N–TiO2 relative to TiO2 and the same was 1.1 fold for UV–visible light exposure. Further, the best catalytic degradation turn-over of the surface determined to be 4.55 10 7 ppm/cm2/s and 1.61 10 7 ppm/cm2/s respectively for UV–vis and visible light exposure. Thus a robust and simpler dip coating methodology giving enhanced visible light photocatalytic activity in polymer mediated nitrogen doped TiO2 system has been demonstrated on a large surface area. We conclude that modification of bandgap of TiO2 by nitrogen doping together with polymer mediated porosity has resulted in a strong visible light activity for this material. Since dip coating approach adopted here is cost-effective and amenable to large scale coating processes, it holds potential in the fabrication of large-scale self-cleansing and antimicrobial surfaces for application in medical and diagnostic industries. & 2016 Elsevier B.V. All rights reserved.
Keywords: Visible light photocatalysis Free radicals Dip coating Photocatalytic surface Doped titanium oxide
1. Introduction Photocatalysis is defined as change in rate of chemical reaction or its initiation by catalyst under illumination. Unlike other chemical reactions, photocatalysis undergoes both oxidation and reduction reactions on same surface under illumination [1]. Current research trend in photocatalysis is on fuel generation, chemical conversion, green synthesis, biomass degradation, industrial effluent treatment, smart surface coating, antimicrobial surfaces, and self-cleansing materials [2,3]. It is pertinent to note that such a broadly investigated field of research has emerged with only very few commercial products applicable for industrial and domestic utility for example Global NANO products INC., Percenta nanotechnology, NanoClear. This is primarily because of inability to be fully utilized in visible region of solar radiation, paucity of data in scale-up systems and field conditions.
n
Corresponding author. E-mail address:
[email protected] (B.M. Jaffar Ali).
http://dx.doi.org/10.1016/j.solmat.2016.03.008 0927-0248/& 2016 Elsevier B.V. All rights reserved.
TiO2, a wide bandgap material existing in different polymorphs, is one of the important catalysts analyzed extensively for its photocatalytic activity in different applications. Still research is being pursued to improve upon TiO2 based end products in many demanding applications [2]. Since bandgap of 3.2 eV lies in UV region of electromagnetic spectrum, utility of TiO2 for commercial applications is limited. In an effort to utilize photons of the visible spectrum by TiO2, its bandgap need to be modified. Generally it is accomplished with metals and non-metals doping [4,5], and by the formation of heterojunctions [4,6]. One of the compelling methods to achieve visible light activity in TiO2 is by non-metal doping, especially placing nitrogen in TiO2 interstitial or substitutional sites [7–9]. In addition to these approaches, minimizing faster recombination through metal based heterojunctions is also pursued that increases generation of free radicals. Recently, following RF magnetron sputtering, fabrication of 15 mm 15 mm surfaces of TiO2 coated on metal sheets have been attempted and validated its photocatalytic effect on microbial system [10]. To make this photocatalyst system commercially viable in any of the emerging applications, it should meet the criterion of low cost and reusability. Commercial level reusability
126
V. Raja et al. / Solar Energy Materials & Solar Cells 152 (2016) 125–132
of TiO2 can be achieved by say, incorporating it in a reusable matrices/substrates. Recent development in magnetic core-shell based system supports magnetic separation for recycling of catalysts [11]. In chemical and biotechnological industries, immobilized catalysts are most commonly used. The ability of a photocatalyst to perform in real time applications can be quantified following series of methodologies or instrumentation based analysis [11–16]. In this work, to partially address the above mentioned concerns, we have fabricated 40 mm 25 mm surfaces of TiO2 and its compounds with lower bandgap, through simple dip coating method. Further to modify bandgap, nitrogen has been incorporated with urea in precursor coating solution. To improve the ability to absorb photons and generate large concentration free radicals, photoactive surface per unit coated surface is maximized by the formation of porous structures on the glass slide. Finally, to quantify the amount of OH* generated, DCFH – a fluorescing probe which emits fluorescence upon activation by OH* radicals, has been analyzed for its applicability for quantifying the photocatalytic ability of a photocatalyst [16,17].
2. Materials and methods 2.1. Chemicals and reagents Tetra ethyl ortho silicate (TEOS), 2-proponal, titanium tetra isopropoxide (TTIP), nitric acid (HNO3) were purchased from Fischer, urea purchased from Finar, PEG-20000 purchased from SRL, methylene blue from Spectrum, and (Di-Chloro Fluorescein) DCFH purchased from Cayman Chemicals Company. All chemicals were used as received without further purification. 2.2. Preparation of precursors for surface coatings 2.2.1. Preparation of precursor solution for silica coating Solution A consists of 5.5 ml of TEOS dissolved in 19.62 ml of 2-proponal, stirred for 5 minutes at room temperature. Solution B consists of 1 ml of HNO3 mixed with 1.25 ml of H2O and 19.25 ml of 2-proponal stirred for 5 min at room temperature. Solution A added to Solution B and stirred for 30 min at room temperature. Resultant solution was used as such for preparation of silica coating over the glass substrate. 2.2.2. Preparation of TiO2 precursor solution To 100 ml of H2O with 1% of HNO3. 9 ml of TTIP added slowly to the solvent under vigorous stirring at room temperature. Stirring continued for 24 h at room temperature, resulting transparent solution. Hydrolyzed solution was condensed at 50 °C for 6 h. 2.2.3. Preparation of N–TiO2 precursor solution While preparing nitrogen doped TiO2, urea as precursor for nitrogen has been utilized. Titanium precursor solution with urea has been taken in 4 different nitrogen to titanium molar ratios, namely 1%, 10%, 20% and 60% in the final solution, referred respectively as N1, N10, N20 and N60. It is prepared by taking 100 ml of H2O containing 1% of HNO3. 9 ml of TTIP added slowly to the solvent under vigorous stirring at room temperature for 24 hours. Thus, hydrolyzed solution was condensed at 50 °C for 6 h. 2.2.4. Preparation of porous-TiO2 precursor Solution Polyethylene glycol (PEG-20,000) has been added in the following three different concentrations: 50 mg, 150 mg, and 500 mg, named as P1, P2, and P3 respectively, prior to preparations of titanium precursor solutions to give p-TiO2 precursor solution.
2.2.5. Preparation of porous-N–TiO2 precursor Solution PEG 150 mg and N 20 concentration of urea has been taken to prepare porous nitrogen doped TiO2 precursor solution as discussed above. Resultant precursor solutions thus prepared were used as such for coating of TiO2 compounds over the Si coated glass slides. 2.3. Dip coating of material over glass slides Indigenous dip coating system with 1.25 mm precision at minimum movement of 6 mm with a load of 2 kg has been assembled with Holmarc (India) linear translator stages and motion controller. 22 ml of prepared coating solutions was taken in 25 ml glass beaker. With a withdrawal speed of 96 mm/min, dip coating of silica solution and Tihydroxide solution has been performed at 25 °C. Silica coating has been performed for one cycle and dried at 100 °C for 15 min. On this dried silica coated slides, 24 cycles of Ti compound precursors have been coated with 15 s immersion period [18] and 5 s hold period outside the solution. This results in 40 mm 25 mm area on the glass slide getting coated. Total transit time per cycle calculated to be 72.5 s. For 24 cycles of coating on a single glass slide, it took about 29 min. Coated slides were air dried for 5 min, followed by heating at the rate of 5 °C/min and held at 150 °C for an hour. It was followed by calcination at 400 °C for three hours. Typical dip coated slides are shown to scale in Fig. S1, (Supplementary Materials). 2.4. Characterization 2.4.1. Spectroscopic characterization Raman scattering of thin films were measured using Renishaw inVia Raman Microscope equipped with 514 nm laser source at 10 mW incident power. UV DRS was carried out in Perkin Elmer spectrometer, Model: Lambda 650, equipped with an integrating sphere attachment. Data were analyzed with BaSO4 as reference. Photoluminescence of thin films were measured by JY Fluorolog-FL3-11 with 450 W Xenon lamp having PMT detection system. Chemical state of catalyst samples analyzed through X-ray Photo electron Spectroscopy model Kratos AXIS ULTRA with Al-kα X-ray (hυ ¼1486.6 eV). Photocatalytic ability of the samples slides has been quantified by tracing the degradation of dyes using custom configured Ocean Optics spectrophotometer having UV–vis light source and USB 2000 detector. Spectrometric data acquisition and analysis were done using SpectraSuite data acquisition system. 2.4.2. Microscopic characterization Scanning Electron Microscope images of the surface coated slides were acquired using Hitachi S-3400N, 15.0 KV secondary electron detection. Atomic Force Microscope images of coated surfaces were scanned in tapping mode using Bruker Multimode 8 enabled with Scanasyst. High resolution Optical Microscope images of coated photoactive surfaces and sections of thin films were acquired using Olympus-BX51 polarizing microscope with UC30-40X objective. 2.4.3. Quantification of photocatalytic activity Dyes exhibit strong adsorption, quantification of their degradation has been very difficult due to camouflaging of adsorption with degradation in the absorption spectra (Fig. S2 Supplementary Material). To circumvent this effect, and to quantify the degradation yield of photocatalyst coated surfaces, semi-continuous flow degradation of a dye molecule (Methylene Blue) has been carried out. 80 W Fluorescent light and 36 W UV light were used to induce photocatalytic activity. In order to include the contribution of photocatalytic activity independent of UV exposure and differentiate the same, a combination of UV light and pure visible light sources are used. Total number of moles of dye molecules degraded over a period of time has been measured from the number of moles of dyes added to the
V. Raja et al. / Solar Energy Materials & Solar Cells 152 (2016) 125–132
system and number of moles remained undegraded in the system as indicated by the decrease in characteristic absorbance of the dye. Two separate measurements were carried out on the sample surface immersed in the dye solution, using UV–visible, referred as LS1, exposed for 7.5 h, whereas, visible light source, referred as LS2, was exposed for 21 hours. 2.4.4. Quantification of free radical generation To one ml of H2O2, 1 ml DCFH was added (40 mM) which make up the final concentration of solution to be 20 mM DCFH. Two of OH radicals convert one molecule of DCFH into DCF* resulting in fluorescence emission at 520 nm upon excitation at 480 nm [17]. A calibration curve was drawn (Fig. S3 in Supplementary Material) for fluorescence count vs concentration of H2O2. Assuming all DCFH at this concentration is converted to DCF*, fluorescence intensity is directly correlated with concentration of OH radicals. To quantify the OH radical generation efficiency of photocatalyst coated slides, freshly coated slides are submerged in water, illuminated with 400 W mercury vapor lamp for 120 min which result in the generation of free radicals. To one ml of photoactivated water, one ml of 40 mM Deacetylated DCFH is added resulting in conversion to DCF* due to the presence of free radicals. Resulting fluorescence emission was quantified for estimating free OH radicals from the calibration curve [17,19,20].
3. Results
3.2. UV–visible absorption spectra To probe the active absorption region of the photocatalytic thin films, UV–visible absorption and Diffuse Reflectance spectra were obtained (data not shown). Fig. 2 gives the Kubelka–Munk function, (F(r).hυ)2 is plotted against hυ, where F(r) is (1 R)2/2R and R is reflectance. h – Planck's constant (J.s); υ – wavelength (nm); 3.3. Photoluminescence Photoluminescence of a semiconductor material provide information about trapping, transfer and migration of charges. Fig. 3 shows the photoluminescence spectra of various thin film samples. 3.4. X-Ray photoelectron spectroscopy To investigate the chemical state of photocatalytic thin film material, XPS analysis has been carried out. Corresponding XPS survey scans of porous-N–TiO2 and porous-TiO2 (0–1200 eV region) are provided in Fig. 4 Inset to Fig. 4 details N 1s peak of both samples. To capture the chemical state of Ti 2p and O 1 s in both samples, a high resolution XPS spectrum around its characteristic peaks are obtained as given Fig. 5. In Fig. 5a and b, Ti3 þ and Ti4 þ states of Ti in porous-N–TiO2 and porous-TiO2 from Ti 2p 3/2 and ½ state respectively has been identified as reported earlier [26]. Fig. 5c and d shows O 1s peak for the respective materials. In
3.1. Raman scattering Raman scattering shifts can differentiate different bonding orbitals that relate to different crystal structures. Hence, Raman shifts has been measured to identify crystal structures of various thin film samples and given in Fig. 1 Raman scattering shifts at 147 cm 1, 197 cm 1, 396 cm 1, 516 cm 1, 638 cm 1 confirms the formation of anatase phase, and 144 cm 1, 238 cm 1, 447 cm 1, 611 cm 1 reveals the presence of rutile phase [21,22]. Low frequency modes at 200 cm 1, 300 cm 1 is in general reported to acoustical phonons, high frequency mode at 575 cm 1 for optical phonons, modes at 330 cm 1, 550 cm 1 is due to first order non-stoichiometric titanium nitride, and 680 cm 1–700 cm 1 broad component ascribed to second order scattering of non-stoichiometric titanium nitride and 460 cm 1, 687 cm 1, 800 cm 1 for defective Ti–N presence [23–25]. Fig. 2. UV-DRS spectra of various TiO2 coated surfaces and their bandgap.
Fig. 1. Raman Scattering for all surface coated TiO2: (i) TiO2 (ii) N- TiO2 (iii) porousTiO2 (iv) porous-N- TiO2.
127
Fig. 3. Photoluminescence spectra of various TiO2 coated surfaces.
128
V. Raja et al. / Solar Energy Materials & Solar Cells 152 (2016) 125–132
N1s spectra 399 eV peak is denoted as N–Ti–O, as reported in previous reports [9]. 3.5. Microscopic analysis To study the surface morphology of the coated slides, Scanning Electron Microscopic analysis has been done for all samples. Fig. 6 gives SEM pictures of photoactive surfaces formed by dip coating. More specifically, SiO2 surface morphology (a), TiO2 (b), porous-TiO2 (c), N–TiO2 (d), porous-N–TiO2 (e) are shown. Representative high resolution SEM image reveal pore formation in polymer doped TiO2 systems (f) suggest a pore diameter of 800 nm. Roughness of photoactive surfaces has been analyzed from AFM images (see Fig. S4, Supplementary Materials) and given in Table 1. Porous samples exhibits more roughness compared to non-porous samples. High resolution optical microscope images of all these surfaces (Fig. S5,
Fig. 4. XPS plot of both porous-TiO2 and porous-N–TiO2 coated glass slides. Inset to Fig. 4 shows high resolution N1s spectrum for same materials.
Supplementary Material) captured and analyzed. Optical microscope images corroborate the appearance of micro cracks in non-porous samples. Also due to the formation of submicron pores in porous samples, crack formation in them appear to be suppressed. Thickness of the films was estimated from the polished sections measured to be in the range of 2–3 mm.
3.6. Photocatalytic degradation Since methylene blue (MB) exhibit high degree of adsorption, in order to differentiate the efficiency of degradation between TiO2 and other compounds, long term, accumulated photo induced degradation of MB by these coated slides has been carried out. Through semicontinuous flow degradation process, photocatalytic activity of surface coated slides has been analyzed. UV–visible light source (LS1) and pure visible light source (LS2) has been used to further delineate the visible light activity of the photocatalytic surfaces. With initial concentration Co, degradation of MB under light sources LS1 has been carried out for 7.5 h. The concentration Co that got depleted due to degradation was restored three times within this time period by adding concentrated stock solution of the MB, and degradation continued. The turn-over rate of photocatalytic degradation of MB in porous-N–TiO2 by LS1and LS2 exposure is determined to be 4.553 10 3 ppm/cm2/s and 1.614 10 3 ppm/cm2/s (Table 1). It is noted that the apparent increase in absolute MB degradation turnover value under UV–vis was due to the use of UV lamp over and above visible light source. Representative absorption spectrum of MB for initial concentration, Co, and after 7.5 h degradation by porous-N– TiO2 is given in Fig. 7. In Fig. 8, performance of various photocatalyst coated surfaces in degrading MB, normalized by exposure time and surface area for light sources LS1 and LS2 are given. Table S1 and S2 (Supplementary Material) summarizes detail of initial and final concentration of MB respectively for photodegradation by LS1 and LS2.
Fig. 5. XPS plot of high resolution spectra of both porous-TiO2 and porous-N–TiO2 coated glass slides: (a) Ti2p of porous-N–TiO2; (b) Ti2p of porous-TiO2; (c) O1s of porousN–TiO2; (d) O1s of porous-TiO2.
V. Raja et al. / Solar Energy Materials & Solar Cells 152 (2016) 125–132
129
Fig. 6. SEM Images of dip coated surfaces: (a) SiO2; (b) TiO2; (c) porous-TiO2; (d) porous-N–TiO2; (e) porous-N–TiO2 ; (f) Representative high resolution image of pore formation in polymer doped TiO2 surface.
3.7. Free radical measurement through DCFH Quantity of free radical generated during photocatalysis is measured utilizing di-chloro fluorescein (DCFH) as a fluorescence reporter molecule. We have determined the OH radical generation in water exposed to 120 minute UV–visible light for all coated surfaces and summarized in Table 1. We found that porous-N– TiO2, surfaces generate 5.69 10 8 ppm/cm2/s of [OH] radicals (Fig. 9).
4. Discussion Predominant presence of anatase phase of TiO2 is confirmed as shown in Fig. 1 by the observation of strong peak at 148 cm 1, and other peaks at 197 cm 1, 402 cm 1, 520 cm 1, 641 cm 1. Also, existence of rutile phase is confirmed from its characteristic peaks at 455 cm 1 and 616 cm 1. It is further noted that presence of
defective Ti–N and non-stoichiometric titanium nitride, explains the modulation of observed bandgap in the system. Weak and broad modes at 337 cm 1 and 565 cm 1 were due to first order non-stoichiometric titanium nitride. It is also noted that Raman scattering due to optical phonon also occurs at 565 cm 1 which make this peak little more prominent as seen in the resolved Raman shifts of Fig. 1 (resolved data not shown). From Kubelka–munk function given in Fig. 2, bandgap for various material coatings have been deduced and presented in Table 1. It is seen that N–TiO2 photocatalyst exhibit red shift of 0.64 eV from its conventional bandgap of 3.2 eV for TiO2 as reported in literature [1,8,27,28]. However we observed 3.08 eV for TiO2 coated slides which may be due to the presence of significant amount of rutile phase[9,26,29]. Further, reduced bandgap observed for N–TiO2 is attributed to oxygen vacancies and nitrogen induced surface states. This fact is corroborated by the observed Raman shift in Fig. 1. In principle our thin film data can be modeled following Bruggeman effective medium approximation [30].
130
V. Raja et al. / Solar Energy Materials & Solar Cells 152 (2016) 125–132
Table 1 Comparison of bandgap of photocatalytic material coatings with surface roughness, methylene blue removal and [OH] radical turn-over rate. Photoactive Surface
TiO2 porous-TiO2 N–TiO2 porous-N– TiO2
Bandgap (eV) Surface Roughness (nm)
3.08 3.04 2.56 2.84
26.7 55.3 28.2 37.4
Amount of MB degraded under different light sources ( 10 7 ppm/ cm2/s) LS1
LS2
4.09 4.21 4.60 4.55
1.09 1.33 1.40 1.61
OH radicals generated ( 10 8 ppm/ cm2/s)
3.69 4.60 3.95 5.69
Fig. 9. Amount of DCFH generated by photocatalytic surfaces in water samples exposed to light source.
Fig. 7. Absorbance profile of Methylene Blue showing characteristic peak at 665 nm before degradation (black dashed line) and after degradation (red dashed line). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Fig. 8. Degradation Methylene blue normalized to catalytic surface area and time of exposure under LS1 (UV–visible light) and LS2 (Visible light) illumination.
However, we note that the average size of the films are 2–3 mm only as determined from thin film sections in light microscope, which render them to be only 4–6 times the pore diameter. It is emphasized that only above a critical thickness of film, which is 50–100 times its pore diameter, Bruggeman model could be applicable. Below this thickness, the dielectric constant becomes function of both pore size (D) and the film thickness [31,32]. Despite this we have extracted effective dielectric constant from
the Reflectance data at its band edge (data not shown). A plot of porosity vs refractive index is made following Bruggeman model corresponds to unrealistic value of porosity for porous- TiO2 and 11.0% porosity for porous-N–TiO2. However it is emphasized that to perform realistic check on the applicability of Bruggeman model, one should have determined effective dielectric constants for various porosity levels. From Photoluminescence spectra given in Fig. 3, significant emission observed at 368 nm is attributed to direct band emission from conduction band to valence band and the 396 nm peak can be due to indirect transition. It is seen that TiO2 sample provide comparatively lesser direct emission, which may be attributed to reduced amount of direct recombination. Porous-TiO2 sample exhibits relatively higher amount of direct recombination with a peak at 367 nm attributed to reduced recombination at indirect transition (396 nm) by scattering of light at pores [33]. N–TiO2 has shown higher amount of indirect recombination compared to direct recombination, can be due to the formation of oxygen vacancies and nitrogen induced surfaces states. N–TiO2 also exhibits a significant emission at around 420 and 468 nm may be due to surface states emission [29,34,35]. It is noted that porous samples exhibit lower indirect to direct recombination ratio compared to non-porous samples giving credence to the fact enhanced photoactivity due to porous nature of sample. AFM studies of the thin film confirms the presence of increased roughness for porous samples. Bright field images corroborate this finding. (see Fig. S2 and Fig. S3, Supplementary Material). Close examination of surfaces of porous films in both images reveal submicron pore structures. In the XPS plot (Figs. 4 and 5), spin orbit lines of Ti 2p 3/2 and ½ states holds a binding energy of 459 eV and 465 eV. Nitrogen doping of samples modify electron density around Ti atom causing the binding energy to be altered, most preferably to lesser energy. Expectedly, it leads to the formation of another peak at 457.1 eV due to Ti–N as shown in Fig. 5a and b. In contrast undoped sample does not undergo any such modifications. Also, it is noted that in the case of porous TiO2 binding energy does not get altered. In Fig. 5c and d, O1s peaks gives information about nature of oxygen binding in porous-N–TiO2 and porous-TiO2 samples. Ti–O has a binding energy of 529.9 eV and Ti–O–Ti has a peak at 531 eV. Since nitrogen doped samples gives change in electronegativity and formation of nitrogen substitution or interstitial doping, peak at 531 eV shifted to lesser binding energy. It has also formed a peak at 528.7 eV which is due to Ti–O–N, which matches the earlier reports [22,28,36].
V. Raja et al. / Solar Energy Materials & Solar Cells 152 (2016) 125–132
It is seen from Fig. 6 that SiO2 material forms a uniform coating over glass slides. However, TiO2 coated over SiO2 surface exhibit micro cracks. High resolution bright field microscopic images (Fig. S5, Supplementary Materials) reveal the presence of microcracks which exhibit considerable growth and widening of crack boundaries in the case of N–TiO2. In samples of porous-TiO2, there is a disappearance of micro cracks due to the formation of porous structure. It is noted that due to submicron size porosity of polymer matrix samples, these surfaces appear to be continuous at this resolution (Fig. S5 b). Similarly, porous N–TiO2 also show disappearance of micro cracks (Fig. S5 d). A representative snap shot of high resolution SEM image (Fig. 6f) for porous-N–TiO2 shows comparatively denser pore structures with an average pore size of 0.87 70.21 mm. In the study of MB degradation, we have exercised care to differentiate passive adsorption against photo-degradation. It has been observed that sharp time-dependent adsorption is significant in MB over all catalyst material coatings (Fig. S2, Supplementary Material). Passive adsorption effect is circumvented in powder materials by taking conveniently a minimum quantity of photocatalytic material and maximizes the catalysis time. However in our case, the photoactive surface is fixed, cannot be minimized. It is therefore pertinent to explore a novel method to quantify the catalytic activity and compare the performance across different surfaces. We have adopted a semicontinuous method of degradation, wherein 0.4 mmol of MB dye solution is prepared and photo-activated to induce catalysis. The concentration C0 was restored three times under continuous catalysis by adding required amount of concentrated (0.2 mmol) stock solution. The detail comparison of relative performance of these materials is carried out (Table S2 and S3, Supplementary Material) estimating amount of dye degraded normalized by the area of catalytic surface, and time duration of catalysis reaction. porous-N–TiO2 coated catalytic surface stands out better among the four surfaces investigated, giving 4.55 10 7 ppm/cm2/s under UV–vis illumination, whereas 1.61 10 7 ppm/cm2/s was observed for visible light illumination, see Table 1. Though the lower value observed for visible light may be attributed to lower intensity of exposure in comparisons to UV–vis intensity, it is emphasized that Visible light exposed material show 1.5 fold increased activity for porous-N–TiO2 coated surface (Table 1) revealing improved visible light performance in comparison to other materials. The mechanism of photodegradation in our system can be understood as follows. Upon illumination on photocatalytic surfaces, valence band electrons were excited to conduction band. This generates an exciton leading to further redox reaction. Photo-generated electrons in conduction band reduce dissolved oxygen species, resulting in superoxide ions. Such superoxide ions have an oxidation potential to generate OH radicals from H2O. Photo-generated holes in valence band readily oxidize, adsorbed H2O and OH radicals into reactive OH radicals. Such reactive OH radicals degrade the methylene blue dye. It is reported that reactive OH radical dominantly have a contribution to lead degradation process in TiO2 photocatalysis compared to other free radicals like super oxide [1,37]. On the basis of our experimental data, the photodegradation mechanism of organic dye by TiO2 system is proposed and explained by the following steps þ
TiO2 þ hυ-eCB þ hVB
TiO2 ðe Þ þ O2
ðdissolvedÞ -O2
2O2 þ 2H þ -H2 O2 þ O2 H2 O2 þ O2 =e =hυ-reactive OH radicals TiO2 h
þ
þ H2 OðadsorbedÞ -TiO2 þ reactive OH radicals þ H þ
TiO2 h
þ
131
=reactive OH radicals=O2 þ Methylene blueðadsorbedÞ -TiO2 þ degraded compounds of MB
In DCFH based OH radical measurement, we record about 1.54 fold increase in OH radical concentration (Fig. 9) for porous-N–TiO2 in comparison to TiO2. It is noted that the same is 1.25 fold for porous TiO2. This corroborate the increased photo degradation ability under visible light for porous-N–TiO2 as a combined effect of enhanced surface area and bandgap tuning. Though, in case of nitrogen doping, due to higher electronegativity to water compared to undoped material contribute to increased free radical formation, we observe dominant contribution to free radical generation coming from enhanced surface area. Degradation of Methylene blue is mediated by O2 and OH* radicals in a photocatalytic reaction. However the reporter molecule DCFH employed here could detect only OH* radicals. From photoluminescence spectra (Fig. 3) it was observed that both porous samples have demonstrated reduced indirect transitions. The DCFH probe give evidence for porous nature enhances generation of free radicals as seen from increased OHn turn-over rate (Fig. 9). It is to be noted that nitrogen doping of TiO2 has not shown significant increase in OHn turn-over rate (Fig. 9) where as it has exhibited marked increase in photocatalytic degradation of Methylene blue (Fig. 8). We can therefore infer that reduction of indirect recombination in porous photoactive surfaces (Fig. 3) has resulted increased OH* radical generation. Whereas in the case of nitrogen doped surfaces, marginal increase in OH* radical generation (Fig. 9) together with other free radicals generated in the system may be contributing to increased photocatalysis. In the mechanism of degradation of MB, it is reported that about 7–8 OH radicals required for complete mineralization [38]. We have shown that DCFH can quantify the amount of OH radical generation under photoactivity. This can enable one to precisely determine amount of MB that can be optimally degraded by the system. For that matter, system can be optimized to degrade any molecules whose mechanism of degradation is fairly well known. This can help construct a scale up models for degradation of dyes. From Table 1, it is clear that OH radical generated is correlated with the ppm of methylene blue removal. Thus DCFH can also be utilized as one of the reporter of OH* for standardization of photocatalytic ability of different surfaces [38,39].
5. Conclusion It is most essential to validate the methodology of material synthesis with desired properties for large scale application. In this work, we have reported a proof of concept for the development of large visible light photoactive surfaces following ease of use dip coating method. A fluorescence based reporter system for OH radical generation has been successfully used to calibrate the degradation activity. Our results show increase in porosity engineered through PEG matrix has resulted in improved photon scattering ability thus giving a beneficial effect in photocatalytic activity. 1.5 fold increase in OH radical generation and 1.54 fold increase in photo-degradation have been demonstrated in porous-N–TiO2 under visible light. We conclude that porous-N–TiO2 system, surface engineered by PEG and reduction in bandgap, has shown to be potentially useful in harnessing sunlight, and in adopting the same for cost effective large scale dip coating methodologies.
Acknowledgments We acknowledge Pondicherry University's Central Instrumentation Facility for Raman scattering, SEM and Fluorescence measurements;
132
V. Raja et al. / Solar Energy Materials & Solar Cells 152 (2016) 125–132
Centre for Nanoscience and Technology for UV-DRS; Department of Physics for AFM analysis. We thank Dr. K. Ramesh, Indian Institute of Science, Bangalore for XPS measurements. Dr. G. Ramesh Babu and Dr. Rajneesh Bhutani for help with microscopic characterization. L.S. acknowledges MNRE, Govt. of India for Research Fellowship.
Appendix A. Supplementary material Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.solmat.2016.03.008.
References [1] B. Viswanathan, Photoelectrochemistry, Principles and Practices Narosa Publications, 2014. [2] L. Gomathi Devi, R. Kavitha, Review on modified N–TiO2 for green energy applications under UV/visible light: selected results and reaction mechanisms, RSC Adv. 4 (2014) 28265. [3] J.C. Colmenares, R. Luque, Heterogeneous photocatalytic nanomaterials: prospects and challenges in selective transformations of biomass-derived compounds, Chem. Soc. Rev. 43 (2014) 765–778. [4] U.G. Akpan, B.H. Hameed, The advancements in sol-gel method of doped-TiO2 photocatalysts, Appl. Catal. A Gen. 375 (2010) 1–11. [5] A. Petala, D. Tsikritzis, M. Kollia, S. Ladas, S. Kennou, D.I. Kondarides, Synthesis and characterization of N-doped TiO2 photocatalysts with tunable response to solar radiation, Appl. Surf. Sci. 305 (2014) 281–291. [6] A. Zaleska, Doped-TiO2 : a review, Recent Pat. Eng. 2 (2008) 157–164. [7] B. Viswanathan, K.R. Krishanmurthy, Nitrogen incorporation in TiO2: does it make a visible light photo-active material? Int. J. Photoenergy (2012) 1–10ID269654. [8] R. Asahi, T. Morikawa, H. Irie, T. Ohwaki, Nitrogen-doped titanium dioxide as visible-light-sensitive photocatalyst : designs, developments, and prospects, Chem. Rev. 144 (2014) 9824–9852. [9] M. Sathish, B. Viswanathan, R.P. Viswanath, C.S. Gopinath, Synthesis, characterization, electronic structure, and photocatalytic activity of nitrogendoped TiO2 nanocatalyst, Chem. Mater. 17 (2005) 6349–6353. [10] S. Cao, B. Liu, L. Fan, Z. Yue, B. Liu, B. Cao, Highly antibacterial activity of N-doped TiO2 thin films coated on stainless steel brackets under visible light irradiation, Appl. Surf. Sci. 309 (2014) 119–127. [11] W. Su, T. Zhang, L. Li, J. Xing, M. He, Y. Zhong, et al., Synthesis of small yolk– shell Fe3O4@TiO2 nanoparticles with controllable thickness as recyclable photocatalysts, RSC Adv. 4 (2014) 8901–8906. [12] A. Mills, C. Hill, P.K.J. Robertson, Overview of the current ISO tests for photocatalytic materials, J. Photochem. Photobiol. A Chem. 237 (2012) 7–23. [13] A. Mills, M. McGrady, A study of new photocatalyst indicator inks, J. Photochem. Photobiol. A Chem. 193 (2008) 228–236. [14] Y. Kikuchi, K. Sunada, T. Iyoda, K. Hashimoto, A. Fujishima, Photocatalytic bactericidal effect of TiO2 thin films: dynamic view of the active oxygen species responsible for the effect, J. Photochem. Photobiol. A Chem. 106 (1997) 51–56. [15] K. Ishibashi, A. Fujishima, T. Watanabe, K. Hashimoto, Quantum yields of active oxidative species formed on TiO2 photocatalyst, J. Photochem. Photobiol. A Chem. 134 (2000) 139–142. [16] L.M. Liu, P. Crawford, P. Hu, The interaction between adsorbed OH and O2 on TiO2 surfaces, Prog. Surf. Sci. 84 (2009) 155–176. [17] C.P. Lebel, H. Ischiropoulos, S.C. Bondys, Evaluation of the probe 20 , 70 -dichlorofluorescein as an indicator of reactive oxygen species formation and oxidative stress, Chem. Res. Toxicol. 5 (1992) 227–231.
[18] M. Addamo, V. Augugliaro, a Di Paola, E. García-López, V. Loddo, G. Marcì, et al., Photocatalytic thin films of TiO2 formed by a sol-gel process using titanium tetraisopropoxide as the precursor, Thin Solid Films 516 (2008) 3802–3807. [19] E. Gutmann, A. Benke, Pyroelectrocatalytic disinfection using the pyroelectric effect of nano-and microcrystalline LiNbO3 and LiTaO3 particles, J. Phys. Chem. C 116 (2012) 5383–5393. [20] L. Bourré, S. Thibaut, A. Briffaud, N. Rousset, S. Eléouet, Y. Lajat, et al., Indirect detection of photosensitizer ex vivo, J. Photochem. Photobiol. B Biol. 67 (2002) 23–31. [21] S. Hu, A. Wang, X. Li, H. Löwe, Hydrothermal synthesis of well-dispersed ultrafine N-doped TiO2 nanoparticles with enhanced photocatalytic activity under visible light, J. Phys. Chem. Solids 71 (2010) 156–162. [22] V. Etacheri, M.K. Seery, S.J. Hinder, S.C. Pillai, Highly visible light active TiO2 xNx heterojunction photocatalysts, Chem. Mater. 22 (2010) 3843–3853. [23] 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. [24] A. Li Bassi, D. Cattaneo, V. Russo, C.E. Bottani, E. Barborini, T. Mazza, et al., Raman spectroscopy characterization of titania nanoparticles produced by flame pyrolysis: the influence of size and stoichiometry, J. Appl. Phys. 98 (2005) 1–9. [25] L. Samiolo, M. Valigi, D. Gazzoli, R. Amadelli, Photo-electro catalytic oxidation of aromatic alcohols on visible light-absorbing nitrogen-doped TiO2, Electrochim. Acta 55 (2010) 7788–7795. [26] X. Li, P. Liu, Y. Mao, M. Xing, J. Zhang, Preparation of homogeneous nitrogendoped mesoporous TiO2 spheres with enhanced visible-light photocatalysis, Appl. Catal. B Environ. 164 (2015) 352–359. [27] M.A.A.S. Y Xu, The absolute energy positions of conduction and valence bands of selected semiconducting minerals, Am. Mineral. 85 (2000) 543–556. [28] R. Asahi, T. Morikawa, T. Ohwaki, K. Aoki, Y. Taga, Visible-light photocatalysis in nitrogen-doped titanium oxides, Science 293 (2001) 269–271. [29] C. Fàbrega, T. Andreu, F. Güell, J.D. Prades, S. Estradé, J.M. Rebled, et al., Effectiveness of nitrogen incorporation to enhance the photoelectrochemical activity of nanostructured TiO2: NH3 versus H2–N2 annealing, Nanotechnology 22 (2011) 235403. [30] Vladimir Kochergin, H. Foll, Porous Semiconductors Optical Properties and Applications, Eng. Mater. Process, Springer-Verlag, London, 2009. [31] A. Garahan, L. Pilon, J. Yin, I. Saxena, Effective optical properties of absorbing nanoporous and nanocomposite thin films, J. Appl. Phys. 014320 (2007). [32] M.M. Braun, L. Pilon, Effective optical properties of non-absorbing nanoporous thin films, Thin Solid Films 496 (2006) 505–514. [33] J. Liqiang, Q. Yichun, W. Baiqi, L. Shudan, J. Baojiang, Y. Libin, et al., Review of photoluminescence performance of nano-sized semiconductor materials and its relationships with photocatalytic activity, Sol. Energy Mater. Sol. Cells 90 (2006) 1773–1787. [34] Y. Cao, T. He, Y. Chen, Y. Cao, Fabrication of rutile TiO2-Sn/anatase TiO2–N heterostructure and its application in visible-light photocatalysis, J. Phys. Chem. C. 114 (2010) 3627–3633. [35] J.B. Yoo, H.J. Yoo, B.W. Lim, K.H. Lee, M.H. Kim, D. Kang, et al., Controlled synthesis of monodisperse SiO2–TiO2 microspheres with a yolk–shell structure as effective photocatalysts, ChemSusChem 5 (2012) 2334–2340. [36] X. Zhou, F. Peng, H. Wang, H. Yu, J. Yang, Effect of nitrogen-doping temperature on the structure and photocatalytic activity of the B,N-doped TiO2, J. Solid State Chem. 184 (2011) 134–140. [37] A. Fujishima, X. Zhang, D. a Tryk, TiO2 photocatalysis and related surface phenomena, Surf. Sci. Rep. 63 (2008) 515–582. [38] a Houas, Photocatalytic degradation pathway of methylene blue in water, Appl. Catal. B Environ. 31 (2001) 145–157. [39] C.A. Lekelefac, P. Czermak, M. Herrenbauer, Evaluation of photocatalytic active coatings on sintered glass tubes by methylene blue, Int. J. Photoenergy (2013) 1–10ID 614567.
