- Email: [email protected]
Effect of sodium carbonate as an additive on the morphology and photocatalytic activity of TiO2 nanotubes
Materials Research Bulletin 95 (2017) 169–176
Contents lists available at ScienceDirect
Materials Research Bulletin journal homepage: www.elsevier.com/locate/matresbu
Effect of sodium carbonate as an additive on the morphology and photocatalytic activity of TiO2 nanotubes M. Alitabar, H. Yoozbashizadeh* Department of Materials Science and Engineering, Sharif University of Technology, Azadi Ave, P.O. Box 11155-9466, Tehran, Iran
A R T I C L E I N F O
Article history: Received 3 May 2017 Received in revised form 10 July 2017 Accepted 12 July 2017 Available online 22 July 2017 Keywords: Pure TiO2 nanotube arrays Doped TiO2 nanotube arrays Sodium carbonate Band gap energy Photocurrent density
A B S T R A C T
The major purpose of this research is increasing the photocatalytic activity of TiO2 nanotube arrays by doping with sodium and carbon for using in water splitting as photoanode. The synthesized TiO2 nanotubes (TNA) were characterized using FESEM (Field Emission Scanning Electron Microscope), XRD (X-ray Diffraction), DRS (Diffraction Reflection Spectroscopy) and XPS test (X-ray Photoelectron Spectroscopy) analyses. The results of FTIR and XPS confirmed the presence of sodium and carbon in the lattice of TNA as dopants. Moreover, the DRS test showed the decrease in the band gap energy of TNA from 3.20 to 2.88 eV; uv-visible test exhibited extension in the absorption edge of pure TiO2 nanotubes from Uv-light (396 nm) to visible light (488 nm). In addition, the results of photoelectrochemical measurement indicated that the photocurrent density of doped TiO2 nanotubes increased about 5 times than that of the pure compounds. © 2017 Elsevier Ltd. All rights reserved.
1. Introduction In the recent decades, the use of clean fuels such as hydrogen has drawn considerable attentions. There are many methods to produce hydrogen; one of them is chemical electrolysis of water under the sun light which is called water splitting reaction [1–3]. This nomination is due to its compatibility with environment and also its low cost [4]. In this technique a material with suitable photocatalytic activity under the sun light, is required [5,6]. In 1972, Honda and Fujishima proposed the use of TiO2 electrode in water electrolysis process for the first time [7,8]. After that, TiO2 as a suitable photocatalytic material has attracted many attentions, thus, different researches have been conducted to investigate the properties of TiO2 nanotubes up to now [9,10]. Actually, among the different structures of TiO2, the nanotube arrays which are synthesized by electrochemical method are preferred because of the long range order, high aspect ratio, large surface area [11], desirable application in solar light alteration, photoelectrochemical stability and low cost [12,13]. Furthermore, the TiO2 nanotubes have better photocatalytic activity than their powder form [14]. The restriction of using these nanotubes in water splitting process as photoanode is their wide band gap energy (about 3.2 ev
* Corresponding author. E-mail addresses: [email protected], [email protected] (H. Yoozbashizadeh). http://dx.doi.org/10.1016/j.materresbull.2017.07.013 0025-5408/© 2017 Elsevier Ltd. All rights reserved.
for anatase phase and 3 eV for rutile phase). Therefore, their photocatalytic activity is only limited to UV-light region (l < 388 nm) which involves very the TiO2 nanotubes small part of solar spectrum of about 5% [15–18]. Despite, the major part of solar energy (approximately 45%) relates to the visible light [19,20]. In fact, by doping with nonmetal elements such as C, N, F, etc and metal elements such as Fe, Co, Ni, Cu, etc, it is possible to extent their absorption edge to the visible light with affecting their electronic structure [21–23]. Indeed, adding metal or non-metal as dopant individually, has not noticeable effect on the absorption edge and photocatalytic activity of nanotubes whilst simultaneous doping with metal and non-metal as dopants is very effective [24]. In this research, the morphology as well as the photocatalytic activity of doped TiO2 nanotubes was investigated. Doping was performed through the adding of sodium carbonate which causes the doping with sodium as metal and carbon as non-metal. The effect of doping on the extention of absorption edge to uv-visible region was studied. 2. Experimental procedure 2.1. Anodizing process All of the anodization experiments were carried out at room temperature in a two-electrode electrochemical cell. The initial materials were Ti foil with 99.9% purity having thickness of
170
M. Alitabar, H. Yoozbashizadeh / Materials Research Bulletin 95 (2017) 169–176
0.25 mm as anode and graphite thin film with thickness of 1 mm as cathode. Before anodizing, Ti foil was sanded with abrasive papers from 320 to 2500 and polished with solution containing Al2O3 particles (0.3 mm). Then, it was rinsed with acetone and ethanol (volume ratio 1:1), afterwards, it was washed with deionized water and finally dried at ambient temperature. All of the experiments were performed at the constant DC voltage of 50 V. Area ratio of anode to cathode was 1: 1 and distance between them was fixed as 3 cm. The used electrolyte includes ethylene glycol, 3 vol% of deionized water, NH4F (0.2 M) and different concentrations of sodium carbonate as an additive (0.01, 0.02, 0.03 and 0.04 M). 2.2. Characterization of TiO2 nanotube arrays The structure and morphology of arrays were investigated by Xray diffraction (XRD, STOE D-64295, Cu Ka radiation, l = 0.15406 nm at a scanning rate of 10 min1) and Field Emission Scanning Electron Microscopy (FESEM, ZEISS SIGMA VP Germany), equipped with EDS. The mean crystallite size of TiO2 nanotubes was determined from the broadening of XRD lines using Williamson Hall procedure [25,26]. To correct the instrumental errors the powder diffraction pattern of an alumina sample was used [26]. To determine the band gap energy, the Diffuse Reflectance Spectra (DRS, Avaspec-2048-TEC) with Ava lamp DH-S setup and UV–vis (Varian Cary 500 UV–vis-NIR spectrophotometer) with wave length of 300–600 nm were applied. Elemental composition of TiO2 nanotubes was evaluated by XPS (X-ray Photoelectron Spectroscopy) test.
2.3. Photoelectrochemical measurement To study the photoelectrochemical performance of the samples, 3 standard electrodes were used where Pt is the counter electrode, Ag/Agcl is the reference electrode and TiO2 nanotubes synthesized with different concentrations of sodium carbonate are working electrodes. Moreover, the electrolyte contained KOH (1 M) in ethylenglycol (10%wt) solution, under the visible light irradiation
(100 mW/cm2) of at 50 mV. To simulate the solar light, 150W xenon ozone-free lamp was applied. 3. Results and discussion 3.1. The morphology characterization Fig. 1 shows the top and cross section FE-SEM images of TiO2 nanotubes which were synthesized in the pure organic electrolyte and also in the organic electrolyte with sodium carbonate as an additive. As could be seen, after one hour of synthesizing, in the pure electrolyte, the nanotubes with length of 0.95 mm and relatively good surface order have been formed. The growth rate of nanotubes calculated by dividing the length to the synthesis time was estimated as 0.016 mm/min. The cross section image of synthesized nanotubes in the organic electrolyte containing additive indicates the formation of, smooth and highly order nanotubes with no ripple on their wall. According to the calculations, by enhancing the concentration of additive from 0.01 to 0.03 M, the growth rate of nanotubes increases from 0.23 to 0.42 mm/min. Further enhance in the concentration from 0.03 to 0.04 M, results in the reduction of growth rate from 0.42 to 0.24 mm/min. In addition, the walls of nanotubes which were synthesized in this concentration (0.04 M) were destroyed. Thus, it could be said that there is an optimal value for concentration of this additive in organic electrolytes which is about 0.03 M. The effect of different concentrations of sodium carbonate on the characterization of TiO2 nanotubes has been summarized in Table 1. It could be inferred from this Table that by increasing the concentration of this additive from 0.01 to 0.03 M, the growth rate and thickness of the nanotubes increase while the inner diameter decreases. Hence, using this additive not only has not negative effect on the tubular structure of nanotubes but also improves their morphology. By considering these results, the best concentration of sodium carbonate as an additive in organic electrolyte is 0.03 M. Generally, the synthesis process of TiO2 nanotubes includes two reactions: 1) electro chemical etching and (2) chemical dissolution [5] as follow: Ti + 2H2O ! TiO2 + 4H+ + 4e
(1)
Fig. 1. Top-view FESEM images of (a) pure TiO2 nanotubes and doped TiO2 nanotubes derived from addition of sodium carbonate with different concentrations of (b) 0.01, (c) 0.02, (d) 0.03 and (e) 0.04 M, respectively, and (f-k) corresponding cross-section FESEM images.
M. Alitabar, H. Yoozbashizadeh / Materials Research Bulletin 95 (2017) 169–176
171
Table 1 Effect of sodium carbonate on the characterizations of TiO2 nanotubes at different concentration. Sodium Carbonate Concentration (M)
Length (mm)
Thickness (nm)
inner Diameter (nm)
0 0.01 0.02 0.03 0.04
0.95 13.96 18.87 25.41 14.1
9.25 14.25 16.67 38.21 16.95
104.14 96.84 89.54 74.16 91.54
TiO2 + 6F+ 4H+ ! TiF62 + 2H2O
(2)
The pH of electrolyte is effective parameter on the rate of electrochemical etch and the chemical dissolution. By controlling this parameter it is possible to change the growth rate of nanotubes [27]. Increasing the pH of electrolyte leads to increase of the hydrolysis rate instead, decreasing the chemical dissolution results in formation of longer nanotubes [28]. Besides, the best range of pH of electrolyte is between 3 and 5; higher pH values leads to formation of longer nanotubes with undesirable debris [29]. According to the results of Macak et al., it is possible to achieve nanotubes with high-aspect-ratio by changing the pH of electrolyte locally. This means, reducing the pH of the bottom of nanotube pores whilst enhancing the pH at the pore walls and the pore mouth [27,30]. Sodium carbonate as a buffer solution in the organic electrolyte can create the pH gradient which is required for fabricating the nanotubes with longer length. However, increasing the sodium carbonate concentration in the electrolyte is effective on the pH gradient and yields to the enhancement of the growth rate of nanotubes. Further increasing from 0.03 to 0.04 M leads to decrease of the growth rate and destroy of nanotubes walls. This phenomenon illustrates another event which occurs during the anodizing process. In this work, after adding the sodium carbonate as an additive to organic electrolyte following reactions may be take place [2,31,32]: Na2CO3 + 6F + H+ ! [NaF6]5 + CO2 + H2O
(3)
Actually, sodium anion complex which is produced in the electrolyte could move towards the anode electrode. Hence, it is probable that TiO2 nanotube arrays be doped by sodium atoms in the presence of electrical field, schemicaly shows in Fig.2. It could be said, the presence of sodium as dopant in the lattice of nanotubes leads to destroy of nanotube walls. This is ascribed to
the large difference between the ionic radius of sodium and titanium. Fig. 3 exhibits the X-Ray map and EDX analysis of TiO2 nanotubes fabricated in organic electrolyte with sodium carbonate, respectively. Evidently, by increasing the concentration of additive from 0.02 to 0.03 M, the intensity of sodium and carbon peaks enhance. The XRD patterns of pure and doped samples have been shown in Fig. 4. All peaks of the pure sample are similar to those reported in pervious works [34,35]. According to the recent researches [33,34], enhancement of annealing temperature to 600 C leads to collapse of nanotube arrays structure while by reducing it to lower than 550 C the charge transfer and their separation efficiency increase. It should be mentioned that the annealing temperature in this work was selected as 500 C. The samples exhibit almost similar diffraction peaks indicating that use of sodium carbonate doesn’t change the phase structure of TiO2 nanotube. The peaks which are located at 2u = 25.1, 37.4 and 47.8 attribute to corresponding planes of anatase (101), (004) and (200) planes, respectively [35,36]. To investigate, the effect of sodium carbonate addition, the peak with highest intensity (004) which is located at 37.4 was selected. As could be observed, the intensity of the peaks related to the nanotube synthesized in organic electrolyte containing sodium carbonate is lower than those of nanotubes synthesized in pure electrolyte. As it is known, the intensity of diffraction peaks indicates the percent of crystallinity [37], thus it can be concluded that use of sodium carbonate leads to the decrease of crystalinity. Another fact that could be deduced from the XRD patterns is the broadening, FWHM (Full Width Half Maximum), of the main peak of anatase ((004)) which has been increased from 0.1889 to 0.2808 when synthesized in the electrolyte containing sodium carbonate. According to
Fig. 2. Schematic diagram indicates the microstructure of pure and doped synthesized TiO2 nanotubes.
172
M. Alitabar, H. Yoozbashizadeh / Materials Research Bulletin 95 (2017) 169–176
Fig. 3. (1) X-Ray Map analysis of doped TiO2 nanotubes and (2) EDX analysis of a) pure, doped nanotubes with different concentrations of additive b) 0.02 and c) 0.03 M.
using the anatase diffraction peaks. According to the results of Table 2, the parameters (a, c) enhance by increasing the concentration of additive. Actually the change of parameters could be the result of entering larger ions such as sodium in the lattice of TiO2 nanotubes [38]. Furthermore, the shift of the XRD diffraction peaks of doped nanotubes to the lower angles than the pure one exhibits the creation of expansion in the lattice of nanotubes, hence, it could demonstrate the presence of sodium as dopant in the lattice of nanotubes [39,40]. In fact, this strain could be the reason of destroying the wall of nanotubes when the concentration of sodium carbonate is 0.04 M. 3.2. XPS test
Fig. 4. XRD patterns of pure and doped TiO2 nanotubes in organic electrolyte.
Table 2 Effect of concentration of sodium carbonate on the lattice parameter of TiO2 nanotubes. photocatalyst
a (Å)
C (Å)
TiO2 0.01 M Na2CO3-TiO2 0.02 M Na2CO3-TiO2 0.03 M Na2CO3-TiO2
3.7892 3.8036 3.8144 3.8295
9.4803 9.4931 9.5023 9.5112
the Williamson-Hall equation, the mean size of crystallites [25] of TiO2 nanotube was estimated as 92 5 and 77 5 nm for pure and doped samples, respectively. In addition, based on this procedure the strain of structure was calculated as 0.095 and 0.14% for the mentioned samples, respectively. Therefore, by considering these results and FESEM images it can be suggested that using the additive leads to development of some expansion and consequently, strain in the lattice of TiO2 nanotubes [38]. The lattice constant parameters (a,c) for doped nanotubes were calculated
To investigate the chemical state of elements and the surface composition of the synthesized nanotubes, the XPS (X-ray Photoelectron Spectroscopy) analysis was performed. Fig. 5 illustrates the XPS spectra of Ti2p, Na1s, O1s and C1s for doped TiO2 nanotube arrays after annealing at 500 C for 2 hours. In Fig. 5a two peaks could be observed at 458.43 and 464.14 eV which are assigned to Ti2p3/2 and Ti2p1/2, respectively. The distance between them is about 5.71 eV confirming that the main chemical state of Ti in TiO2 nanotube lattice is Ti4+ [41]. A high resolution spectrum of sodium has been illustrated in Fig. 5b. The peak located at 1069.95 eV corresponds to Na1s showing the presence of Na in the lattice of TiO2 nanotubes [42]. Furthermore, according to Fig. 6 the peaks of Ti 2p(3/2) and Ti 2p(1/ 2) for pure nanotubes locate at 459.10 and 464.80 ev, respectively which are in agreement with the literature reported values [43]; these peaks for doped samples shift to the lower values which could be a result of reduction of Ti4+ to Ti3+. In fact, the incorporation of sodium ions in the lattice of nanotubes yields to the creation of oxygen vacancies in the doped samples. Accordingly, the negative charge introduced in the lattice of nanotubes is instead compensated by conversion of Ti4+ to Ti3+ [44]. To confirm this, the high resolution spectrum of oxygen in the doped nanotubes has been investigated as well [5,45]. Fig. 5c shows the high-resolution XPS spectrum of O1s. The O1s peaks locate at 530.50, 531.50 and 533.30 eV which belong to Ti-O bonds (existence of oxygen in crystal lattice of nanotubes), oxygen faults and the hydroxyl groups obtained from chemisorbed water, respectively [46]. The area under the peak situated at 531.5 eV is
M. Alitabar, H. Yoozbashizadeh / Materials Research Bulletin 95 (2017) 169–176
173
Fig. 5. XPS test of doped TiO2 nanotubes. (a) titanium, (b) sodium, (c) oxygen and (d) carbon.
Fig. 6. XPS test of Ti2p1/2, Ti2p3/2 for pure and doped TiO2 nanotubes.
sodium ions inside the lattice of TiO2 nanotubes (not on their surface) could be proved. Fig. 5d illustrates the XPS spectra of TiO2 nanotubes doped with C. As shown in this figure, five peaks, one strong peak at bonding energy of 284.50 eV and four weak peaks at bonding energies of 282.47, 286.47, 287.90 and 288.97 eV could be distinguished. The peak observed around 284.50 eV is assigned to Carbon absorption on the surface as contamination (C C bonds). The peaks at 287.90 and 288.97 eV are attributed to C—O bond (hydroxyl composition) and C¼O bonds (Carboxylic groups), respectively. The observed peaks situating at 286.47 and 282.87 eV are ascribed to Ti-C O as carbonate species and Ti-C bonds, respectively [47]. The carbonate species could be formed as carbon atoms diffuse into the interstitial locations of the TiO2 lattice. However, the presence of C atoms inside the TiO2 nanotubes lattice could not be proved. To confirm the presence of carbon in the lattice of TiO2 nanotube, the existence of a peak in the range of 281.5 283 eV is necessary [5,6]. Here, a very small peak detected at 282.47 eV which is related to TiC bonds in the nanotube lattice indicates the low amount of doped carbon in TiO2 nanotubes. The origin of doped carbon could be related to the ethylenhlycol which exists in the organic electrolyte. It should be mentioned that the results of XRD and FE-SEM analyses also confirmed this fact. 3.3. Uv-visible tests
important parameter, since it is ascribed to the oxygen vacancy [45], which can be developed during the metal and none metal doping process in the lattices of TiO2 nanotubes [47]. Finally, using the results of XPS, XRD and FESEM analyses, the presence of
Fig. 7 exhibits the absorption edge (in the range of 250–690 nm) and the band gap energy of pure and doped nanotubes. It could be seen in Fig.7a that the strong absorption was occurred in the wavelength range of almost uv-light (l < 400 nm) and the weak
174
M. Alitabar, H. Yoozbashizadeh / Materials Research Bulletin 95 (2017) 169–176
Fig. 7. a) uv-visible tests and b) DRS plots of pure and doped TiO2 nanotubes.
absorption was happened in the visible region. It could be said that the strong absorption is due to intrinsic inter-band transition while the weak absorption can be assigned to the forming of faults depending on the synthesis method [48]. As could be observed, in the plots of doped nanotubes, all of samples exhibit stronger absorption in the range of 400–690 nm. Comparing the results of doped nanotubes shows the enhancing of absorption by increasing the additive concentration from 0.02 to 0.03 M. However, among of the doped samples the largest optical absorption edge belongs to the sample synthesized in the electrolyte containing 0.03 M of additive. Furthermore, inserting the additive shifts the absorption edge of pure nanotubes by about 92 nm to the longer wavelength (from 396 to 488 nm). To determine the effect of additive on the band gap energy of TiO2 nanotubes, DRS (Diffuse Reflectance Spectra) analysis was applied which is indicated in Fig. 7b. The band gap energy was calculated using Kubak Munk function and Tauc plot. Based on the Kubak Munk function, a= (1-R)2/(2R), where R is diffuse reflection. In Tauc plot (ahv) 0.5 is drawn versus hv, where h is the Pelank constant. Actually, by extrapolating the linier portion of (ahv) 0.5 graph to the zero value, the band gap energy could be estimated [49]. As Fig. 7b illustrates, the band gap energy of TiO2 nanotubes without any additives is equal to 3.20 eV for anatase phase. Clearly, the use of sodium carbonate reduced the band gap energy of TiO2 nanotubes from 3.20 to 2.88 eV. This reduction could be related to the presence of sodium and carbon in the lattice of TiO2 nanotubes. It seems that the sodium develops some trap levels between valance and conductive bands close to the conduction band. Therefore, it induces the excitation of electrons from the valance band to the trap levels [40,50]. In addition, mixing the 2p state of carbon with 2p state of oxygen creates new valance band upwards the valance band of the anatase phase [51,52].
lower potential is needed to separate the photoinduced electrons and inject them to the outer circuit in doped TiO2 nanotubes [51– 54] Generally, doping process on the TiO2 nanotube arrays leads to the formation of oxygen vacancy and consequently, the creation of unpaired electrons or Ti3+ centers to keep the charge balance. These vacancy states could act as the trapper center of electrons. sodium ions are homogenously dispersed [55], since they could capture the electrons and holes and thus, retard the recombination of electron-hole pairs, thereby, the photocatalytic activity enhances [56]. Furthermore, the substitution of oxygen with carbon atoms in TiO2 lattice leads to the mixture of 2p states of carbon with 2p states of oxygen and development of the new valance band (VB) [16,57], hence, the VB edge of doped TNA moves upwards and the electrons in 2p states of carbon can excited with visible light. In other words, the recombination rate of electron-hole pairs decreases in the doped TiO2 [33]. This implies that more photoelectrons-hole pairs have been generated under the radiation of solar light which could participate in photocatalytic reactions. Overally, the photocatalytic activity of a material means its ability to create electron-hole pairs [50,55]. Actually, the rate of generating of charge carries, which transfer from interior to the surface of photoanode, should be higher than the rate of recombination of photo induced electrons and holes [40,41].
3.4. Effect of additive on the photocurrent density The photocurrent measurement has been utilized in order to determine the convertion ability of photons to current in doped TiO2 nanotubes. Fig. 8 displays the linear sweep voltammograms (LSV) for different samples under visible light irradiation. As it is seen, in the constant applied potentials, the photocurrent density of all doped TiO2 nanotube arrays is significantly higher than that of pure TiO2 nanotube array. The shift of zero current potential to the more negative values in the doped samples comparing to the pure TiO2 nanotubes (from 0.20 V for pure nanotubes to 0.58 V for doped nanotubes) may be related to the combined effect of Na and C atoms in the lattice of doped nanotubes. This means that the
Fig. 8. Photocurrentdensity as a functional of potential for pure and doped TiO2 nanotubes.
M. Alitabar, H. Yoozbashizadeh / Materials Research Bulletin 95 (2017) 169–176
175
photocatalytic activity of nanotubes enhanced. The photocurrent density in doped nanotubes was about 5 times higher than that of pure one indicating the positive effect of sodium Carbonate. Acknowledgment The authors are grateful to the extracting and corrosion laboratories of Sharif University of Technology. Moreover, we would like to give our special thanks to Mr. M. Farokhgisoo and Mrs. Familifard. Appendix A. Supplementary data
Fig. 9. Photocurrent density as a function of time for pure and doped TiO2 nanotubes.
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j. materresbull.2017.07.013. References
According to the results, with doping the TiO2 nanotube arrays with sodium and carbon, the recombination rate of electron-hole pairs reduces and the density of charge carrier existing on the surface of photoanode increases which accelerates the water splitting reactions on its surface [50]. Fig. 9 displays the photocurrent density of pure and doped nanotubes (0.03 M) as a dependence of time in the electrolyte containing KOH and 10%wt ethylenglycol under the visible light irradiation (100 mw/cm2) at 0.5 V. As.Ag/AgCl [49] it is observed that the photocurrent density of doped samples is more than 5 times than of that of pure sample, which shows the positive effect of additive, this is in agreement with photocurrent density-voltage results. Many factors can improve the photocatalytic activity of nanotubes which are synthesized in the organic electrolyte containing sodium carbonate as an additive. According to the results, by irradiation the energy equal or higher than the band gap energy of TiO2 nanotubes, it is possible to generate the electrons and holes in the conduction band (CB) or valance band (VB), respectively. Reducing the band gap energy by doping sodium and carbon could reduce the energy needed for exciting the electrons and holes. Furthermore, sodium and carbon act as trapper center for electron and holes yielding to the reduction of recombination rate and increasing the photocatalytic activity of nanotubes, thus, the excitation range of electron-hole pairs enhances from uv-light to uv-visible range. Obviously, the length and crystallite size of nanotubes is effective parameter on the photocatalytic activity. It has been shown that by increasing the length and decreasing the crystallite size [58] of nanotube the photocatalytic activity enhances. 4. Conclusion In this work, anodizing method was used to synthesize the TiO2 nanotubes in organic electrolyte. When the nanotubes are applied as photoanode in water splitting process, it is necessary to reduce the band gap energy and enhance the photocatalytic activity of nanotubes. In this research, the anodizing time and band gap energy of TiO2 nanotubes reduced simultaneously by using sodium carbonate as an additive. The use of this additive led to development sodium and carbon co-doped nanotube arrays, accordingly, the band gap energy reduced. The results revealed the development of Ti-C, Ti-C O bonds, the existence of a peak at 1072.5 eV confirmed the presence of sodium and Carbon in nanotubes lattice. Furthermore, using this additive increased the separation rate of electrons and holes. Moreover, the
[1] O.K. Varghese, M. Paulose, K. Shankar, G.K. Mor, C.A. Grimes, Nanosci. Nanotechnol. 5 (2005) 1158. [2] Y. Li, Y. Xiang, S. Peng, X. Wang, L. Zhou, Electrochim. Acta 87 (2013) 794–800. [3] R. Beranek, J.M. Macak, M. Gärtner, K. Meyer, P. Schmuki, Electrochim. Acta 54 (2009) 2640–2646. [4] H. Yan, X. Wang, M. Yao, X. Yao, Prog. Nat. Sci. Mater. Int. 23 (2013) 402–407. [5] M. Mollavali, C. Falamaki, S. Rohani, Int. J. Hydrogen Energy 40 (2015) 12239– 12252. [6] M. Mollavali, C. Falamaki, S. Rohani, Int. J. Hydrogen Energy 41 (2016) 5887– 5901. [7] A. Fujishima, K. Honda, Nature 238 (1972) 37–38. [8] Angelis De Filippo, I. Fantacc, Selloni Annabella, K. Mohammad, M. Grätzelx, J. Phys. Chem. C 114 (2010) 6054–6061. [9] Kesong Yang, Ying Dai, Baibiao Huang, S. Han, Phys. Chem. B 110 (2006) 24011– 24014. [10] S. Li, S. Lin, J. Liao, N. Pan, D. Li, J. Li, Int. J. Photoenergy 2012 (2012) 1155. [11] L. Wenjun, Y. Tao, W. Yaoming, L. Zhaosheng, Y. Jinhua, Z. Zhigang, J. Phys. D: Appl. Phys. 40 (2007) 1091. [12] T. Umebayashi, T. Yamaki, H. Itoh, K. Asai, J. Phys. Chem. Solids 63 (2002) 1909– 1920. [13] D. Regonini, C.R. Bowen, A. Jaroenworaluck, R. Stevens, Mater. Sci. Eng. R: Rep. 74 (2013) 377–406. [14] J. Xing, Z. Xia, J. Hu, Y. Zhang, L. Zhong, Corros. Sci. 75 (2013) 212–219. [15] J.P. Xu, J.F. Wang, Y.B. Lin, X.C. Liu, Z.L. Lu, Z.H. Lu, L.Y. Lv, F.M. Zhang, Y.W. Du, J. Phys. D Appli. Phys. 40 (2007) 4757. [16] Xiaobo Chen, C. Burda, J. Am. Chem. Soc. 130 (2008) 5018. [17] S. Karthik, T. Kong Chhay, K.M. Gopal, A.G. Craig, J. Phys. D Appl. Phys. 39 (2006) 2361. [18] Z. Zhao, Q. Liu, J. Phys D Appl Phys 41 (2008). [19] M. Guo, J. Du, Phys. B Condens. Matter. B 407 (2012) 1003–1007. [20] Q. Liu, D. Ding, C. Ning, X. Wang, Int. J. Hydrogen Energy 40 (2015) 2107–2114. [21] H. Liu, G. Liu, X. Shi, Coll. Surf. A 363 (2010) 35–40. [22] A.B. Murphy, Sol. Energy Mater. Sol. Cells 91 (2007) 1326–1337. [23] V.A. Tran, T.T. Truong, T.A.P. Phan, T.N. Nguyen, T.V. Huynh, A. Agresti, S. Pescetelli, T.K. Le, A. Di Carlo, T. Lund, S.-N. Le, P.T. Nguyen, Appl. Surf. Sci. 399 (2017) 515–522. [24] O. Ola, M.M. Maroto-Valer, J. Photochem. Photobiol. C Photochem. Rev. 24 (2015) 16–42. [25] W.H. Hall, G.K. Williamson, J. Sci. Instrum. 29 (1952) 132. [26] T. Gheiratmand, H.R.M. Hosseini, P. Davami, F. Ostadhossein, M. Song, M. Gjoka, Nanoscale 5 (2013) 7520–7527. [27] S. Rani, S.C. Roy, M. Paulose, O.K. Varghese, G.K. Mor, S. Kim, S. Yoriya, T.J. LaTempa, C.A. Grimes, Phys. Chem. Chem. Phys. 12 (2010) 2780–2800. [28] K. Lee, A. Mazare, P. Schmuki, Chem. Rev. 114 (2014) 9385–9454. [29] D.P. Padiyan, D.H. Raja, Energy Procedia 22 (2012) 88–100. [30] J.M. Macak, P. Schmuki, Electrochim. Acta 52 (2006) 1258–1264. [31] Baiqi Shao, Song Yan, Mengmeng Jiao, Wei Lü, Hongpeng You, Inorg. Chem. 55 (2016) 1912–1919. [32] A.S. Povarennykh, Crystal Chemical Classification of Minerals, Springer, Ukrainia, 1972. [33] J. Tian, H. Gao, H. Deng, L. Sun, H. Kong, P. Yang, J. Chu, J. Alloys Compd. 581 (2013) 318–323. [34] J.F. Moulder, F. Stickle, E. Sobol, D. Bomben, Handbook of X-ray Photoelectron Spectroscopy, United State of America, 1992. [35] Y. Liao, W. Que, Q. Jia, Y. He, J. Zhang, P. Zhong, J. Mater. Chem. 22 (2012) 7937– 7944. [36] Yulong Liao, Huaiwu Zhang, Jie Li, Guoliang Yu, Zhiyong Zhong, Feiming Bai, Lijun Jia, Shihong Zhang, P. Zhong, J. Appl. Phys. 115 (2014) 304-. [37] F. Marzpour Shalmani, R. Halladj, S. Askari, Ultrason. Sonochem. 29 (2016) 354–362.
176
M. Alitabar, H. Yoozbashizadeh / Materials Research Bulletin 95 (2017) 169–176
[38] Anderson Janotti, Bharat Jalan, Susanne Stemmer, C.G.V.d. Walle, Appl. Phys. Lett. 100 (2012) (262104-262101-262104-262103). [39] W. Li, J. Yang, J. Zhang, S. Gao, Y. Luo, M. Liu, Mater. Res. Bull. 57 (2014) 177–183. [40] S. Shalini, N. Prabavathy, R. Balasundaraprabhu, T. Satish Kumar, P. Walke, S. Prasanna, D. Velayuthapillai, J. Mater. Sci. Mater. Electron. (2016) 1–9. [41] L.G. Devi, R. Kavitha, Appl. Catal. B Environ. 140–141 (2013) 559–587. [42] T.G. Deepak, D. Subash, G.S. Anjusree, K.R.N. Pai, S.V. Nair, A.S. Nair*, ACS Sustainable Chem. Eng. 2 (2014) 2772–2780. [43] M. Nischk, P. Mazierski, Z. Wei, K. Siuzdak, N.A. Kouame, E. Kowalska, H. Remita, A. Zaleska-Medynska, Appl. Surf. Sci. 387 (2016) 89–102. [44] S. Munir, S.M. Shah, H. Hussain, R. Ali khan, Mater. Des. 92 (2016) 64–72. [45] W. Zhongbiao, D. Fan, Z. Weirong, W. Haiqiang, L. Yue, G. Baohong, Nanotechnology 20 (2009) 235701. [46] N. Yan, Z. Zhu, J. Zhang, Z. Zhao, Q. Liu, Mater. Res. Bull. 47 (2012) 1869–1873. [47] S. Sakthivel, H. Kisch, Angew. Chem. Int. Ed. 42 (2003) 4908–4911. [48] S. Zhang, F. Peng, H. Wang, H. Yu, S. Zhang, J. Yang, H. Zhao, Catal. Commun. 12 (2011) 689–693.
[49] H. Ali, N. Ismail, M. Mekewi, A.C. Hengazy, J. Solid State Electrochem. 19 (2015) 3019–3026. [50] N.L. Huan Xie, Baoshun Liu, Jingjing Yang, Xiujian Zhao, J. Phys. Chem. C 120 (2016) 10390–10399. [51] Y.L. Pang, S. Lim, H.C. Ong, W.T. Chong, Appl. Catal. A Gen. 481 (2014) 127–142. [52] R. Yuan, B. Zhou, D. Hua, C. Shi, Front. Environ. Sci. Eng. 9 (2015) 850–860. [53] R. Asahi, T. Morikawa, T. Ohwaki, K. Aoki, Y. Taga, Science 293 (2001) 269–271. [54] Xiaoping Han, G. Shao*, J. Phys. Chem. C 115 (2011) 8274–8282. [55] M. Zhou, J. Yu, J. Hazard. Mater. 152 (2008) 1229–1236. [56] Zheshuai Lin, Alexander Orlov, Richard M. Lambert, M.C. Payne, J. Phys, Chem. B 109 (2005) 20948–20952. [57] F. Esaka, K. Furuya, J. Vac. Sci. Technol. A Vac. Surf. Films 15 (1998). [58] Mahamasuhaimi Masae, Pichaya Pitsuwan, Chaturong Pholthawon, Niti Pawanwatcharakorn, Lek Sikong, P. Kongsong, Thammasat Int. J. Sci. Technol. 20 (2015) 63–71.
Contents lists available at ScienceDirect
Materials Research Bulletin journal homepage: www.elsevier.com/locate/matresbu
Effect of sodium carbonate as an additive on the morphology and photocatalytic activity of TiO2 nanotubes M. Alitabar, H. Yoozbashizadeh* Department of Materials Science and Engineering, Sharif University of Technology, Azadi Ave, P.O. Box 11155-9466, Tehran, Iran
A R T I C L E I N F O
Article history: Received 3 May 2017 Received in revised form 10 July 2017 Accepted 12 July 2017 Available online 22 July 2017 Keywords: Pure TiO2 nanotube arrays Doped TiO2 nanotube arrays Sodium carbonate Band gap energy Photocurrent density
A B S T R A C T
The major purpose of this research is increasing the photocatalytic activity of TiO2 nanotube arrays by doping with sodium and carbon for using in water splitting as photoanode. The synthesized TiO2 nanotubes (TNA) were characterized using FESEM (Field Emission Scanning Electron Microscope), XRD (X-ray Diffraction), DRS (Diffraction Reflection Spectroscopy) and XPS test (X-ray Photoelectron Spectroscopy) analyses. The results of FTIR and XPS confirmed the presence of sodium and carbon in the lattice of TNA as dopants. Moreover, the DRS test showed the decrease in the band gap energy of TNA from 3.20 to 2.88 eV; uv-visible test exhibited extension in the absorption edge of pure TiO2 nanotubes from Uv-light (396 nm) to visible light (488 nm). In addition, the results of photoelectrochemical measurement indicated that the photocurrent density of doped TiO2 nanotubes increased about 5 times than that of the pure compounds. © 2017 Elsevier Ltd. All rights reserved.
1. Introduction In the recent decades, the use of clean fuels such as hydrogen has drawn considerable attentions. There are many methods to produce hydrogen; one of them is chemical electrolysis of water under the sun light which is called water splitting reaction [1–3]. This nomination is due to its compatibility with environment and also its low cost [4]. In this technique a material with suitable photocatalytic activity under the sun light, is required [5,6]. In 1972, Honda and Fujishima proposed the use of TiO2 electrode in water electrolysis process for the first time [7,8]. After that, TiO2 as a suitable photocatalytic material has attracted many attentions, thus, different researches have been conducted to investigate the properties of TiO2 nanotubes up to now [9,10]. Actually, among the different structures of TiO2, the nanotube arrays which are synthesized by electrochemical method are preferred because of the long range order, high aspect ratio, large surface area [11], desirable application in solar light alteration, photoelectrochemical stability and low cost [12,13]. Furthermore, the TiO2 nanotubes have better photocatalytic activity than their powder form [14]. The restriction of using these nanotubes in water splitting process as photoanode is their wide band gap energy (about 3.2 ev
* Corresponding author. E-mail addresses: [email protected], [email protected] (H. Yoozbashizadeh). http://dx.doi.org/10.1016/j.materresbull.2017.07.013 0025-5408/© 2017 Elsevier Ltd. All rights reserved.
for anatase phase and 3 eV for rutile phase). Therefore, their photocatalytic activity is only limited to UV-light region (l < 388 nm) which involves very the TiO2 nanotubes small part of solar spectrum of about 5% [15–18]. Despite, the major part of solar energy (approximately 45%) relates to the visible light [19,20]. In fact, by doping with nonmetal elements such as C, N, F, etc and metal elements such as Fe, Co, Ni, Cu, etc, it is possible to extent their absorption edge to the visible light with affecting their electronic structure [21–23]. Indeed, adding metal or non-metal as dopant individually, has not noticeable effect on the absorption edge and photocatalytic activity of nanotubes whilst simultaneous doping with metal and non-metal as dopants is very effective [24]. In this research, the morphology as well as the photocatalytic activity of doped TiO2 nanotubes was investigated. Doping was performed through the adding of sodium carbonate which causes the doping with sodium as metal and carbon as non-metal. The effect of doping on the extention of absorption edge to uv-visible region was studied. 2. Experimental procedure 2.1. Anodizing process All of the anodization experiments were carried out at room temperature in a two-electrode electrochemical cell. The initial materials were Ti foil with 99.9% purity having thickness of
170
M. Alitabar, H. Yoozbashizadeh / Materials Research Bulletin 95 (2017) 169–176
0.25 mm as anode and graphite thin film with thickness of 1 mm as cathode. Before anodizing, Ti foil was sanded with abrasive papers from 320 to 2500 and polished with solution containing Al2O3 particles (0.3 mm). Then, it was rinsed with acetone and ethanol (volume ratio 1:1), afterwards, it was washed with deionized water and finally dried at ambient temperature. All of the experiments were performed at the constant DC voltage of 50 V. Area ratio of anode to cathode was 1: 1 and distance between them was fixed as 3 cm. The used electrolyte includes ethylene glycol, 3 vol% of deionized water, NH4F (0.2 M) and different concentrations of sodium carbonate as an additive (0.01, 0.02, 0.03 and 0.04 M). 2.2. Characterization of TiO2 nanotube arrays The structure and morphology of arrays were investigated by Xray diffraction (XRD, STOE D-64295, Cu Ka radiation, l = 0.15406 nm at a scanning rate of 10 min1) and Field Emission Scanning Electron Microscopy (FESEM, ZEISS SIGMA VP Germany), equipped with EDS. The mean crystallite size of TiO2 nanotubes was determined from the broadening of XRD lines using Williamson Hall procedure [25,26]. To correct the instrumental errors the powder diffraction pattern of an alumina sample was used [26]. To determine the band gap energy, the Diffuse Reflectance Spectra (DRS, Avaspec-2048-TEC) with Ava lamp DH-S setup and UV–vis (Varian Cary 500 UV–vis-NIR spectrophotometer) with wave length of 300–600 nm were applied. Elemental composition of TiO2 nanotubes was evaluated by XPS (X-ray Photoelectron Spectroscopy) test.
2.3. Photoelectrochemical measurement To study the photoelectrochemical performance of the samples, 3 standard electrodes were used where Pt is the counter electrode, Ag/Agcl is the reference electrode and TiO2 nanotubes synthesized with different concentrations of sodium carbonate are working electrodes. Moreover, the electrolyte contained KOH (1 M) in ethylenglycol (10%wt) solution, under the visible light irradiation
(100 mW/cm2) of at 50 mV. To simulate the solar light, 150W xenon ozone-free lamp was applied. 3. Results and discussion 3.1. The morphology characterization Fig. 1 shows the top and cross section FE-SEM images of TiO2 nanotubes which were synthesized in the pure organic electrolyte and also in the organic electrolyte with sodium carbonate as an additive. As could be seen, after one hour of synthesizing, in the pure electrolyte, the nanotubes with length of 0.95 mm and relatively good surface order have been formed. The growth rate of nanotubes calculated by dividing the length to the synthesis time was estimated as 0.016 mm/min. The cross section image of synthesized nanotubes in the organic electrolyte containing additive indicates the formation of, smooth and highly order nanotubes with no ripple on their wall. According to the calculations, by enhancing the concentration of additive from 0.01 to 0.03 M, the growth rate of nanotubes increases from 0.23 to 0.42 mm/min. Further enhance in the concentration from 0.03 to 0.04 M, results in the reduction of growth rate from 0.42 to 0.24 mm/min. In addition, the walls of nanotubes which were synthesized in this concentration (0.04 M) were destroyed. Thus, it could be said that there is an optimal value for concentration of this additive in organic electrolytes which is about 0.03 M. The effect of different concentrations of sodium carbonate on the characterization of TiO2 nanotubes has been summarized in Table 1. It could be inferred from this Table that by increasing the concentration of this additive from 0.01 to 0.03 M, the growth rate and thickness of the nanotubes increase while the inner diameter decreases. Hence, using this additive not only has not negative effect on the tubular structure of nanotubes but also improves their morphology. By considering these results, the best concentration of sodium carbonate as an additive in organic electrolyte is 0.03 M. Generally, the synthesis process of TiO2 nanotubes includes two reactions: 1) electro chemical etching and (2) chemical dissolution [5] as follow: Ti + 2H2O ! TiO2 + 4H+ + 4e
(1)
Fig. 1. Top-view FESEM images of (a) pure TiO2 nanotubes and doped TiO2 nanotubes derived from addition of sodium carbonate with different concentrations of (b) 0.01, (c) 0.02, (d) 0.03 and (e) 0.04 M, respectively, and (f-k) corresponding cross-section FESEM images.
M. Alitabar, H. Yoozbashizadeh / Materials Research Bulletin 95 (2017) 169–176
171
Table 1 Effect of sodium carbonate on the characterizations of TiO2 nanotubes at different concentration. Sodium Carbonate Concentration (M)
Length (mm)
Thickness (nm)
inner Diameter (nm)
0 0.01 0.02 0.03 0.04
0.95 13.96 18.87 25.41 14.1
9.25 14.25 16.67 38.21 16.95
104.14 96.84 89.54 74.16 91.54
TiO2 + 6F+ 4H+ ! TiF62 + 2H2O
(2)
The pH of electrolyte is effective parameter on the rate of electrochemical etch and the chemical dissolution. By controlling this parameter it is possible to change the growth rate of nanotubes [27]. Increasing the pH of electrolyte leads to increase of the hydrolysis rate instead, decreasing the chemical dissolution results in formation of longer nanotubes [28]. Besides, the best range of pH of electrolyte is between 3 and 5; higher pH values leads to formation of longer nanotubes with undesirable debris [29]. According to the results of Macak et al., it is possible to achieve nanotubes with high-aspect-ratio by changing the pH of electrolyte locally. This means, reducing the pH of the bottom of nanotube pores whilst enhancing the pH at the pore walls and the pore mouth [27,30]. Sodium carbonate as a buffer solution in the organic electrolyte can create the pH gradient which is required for fabricating the nanotubes with longer length. However, increasing the sodium carbonate concentration in the electrolyte is effective on the pH gradient and yields to the enhancement of the growth rate of nanotubes. Further increasing from 0.03 to 0.04 M leads to decrease of the growth rate and destroy of nanotubes walls. This phenomenon illustrates another event which occurs during the anodizing process. In this work, after adding the sodium carbonate as an additive to organic electrolyte following reactions may be take place [2,31,32]: Na2CO3 + 6F + H+ ! [NaF6]5 + CO2 + H2O
(3)
Actually, sodium anion complex which is produced in the electrolyte could move towards the anode electrode. Hence, it is probable that TiO2 nanotube arrays be doped by sodium atoms in the presence of electrical field, schemicaly shows in Fig.2. It could be said, the presence of sodium as dopant in the lattice of nanotubes leads to destroy of nanotube walls. This is ascribed to
the large difference between the ionic radius of sodium and titanium. Fig. 3 exhibits the X-Ray map and EDX analysis of TiO2 nanotubes fabricated in organic electrolyte with sodium carbonate, respectively. Evidently, by increasing the concentration of additive from 0.02 to 0.03 M, the intensity of sodium and carbon peaks enhance. The XRD patterns of pure and doped samples have been shown in Fig. 4. All peaks of the pure sample are similar to those reported in pervious works [34,35]. According to the recent researches [33,34], enhancement of annealing temperature to 600 C leads to collapse of nanotube arrays structure while by reducing it to lower than 550 C the charge transfer and their separation efficiency increase. It should be mentioned that the annealing temperature in this work was selected as 500 C. The samples exhibit almost similar diffraction peaks indicating that use of sodium carbonate doesn’t change the phase structure of TiO2 nanotube. The peaks which are located at 2u = 25.1, 37.4 and 47.8 attribute to corresponding planes of anatase (101), (004) and (200) planes, respectively [35,36]. To investigate, the effect of sodium carbonate addition, the peak with highest intensity (004) which is located at 37.4 was selected. As could be observed, the intensity of the peaks related to the nanotube synthesized in organic electrolyte containing sodium carbonate is lower than those of nanotubes synthesized in pure electrolyte. As it is known, the intensity of diffraction peaks indicates the percent of crystallinity [37], thus it can be concluded that use of sodium carbonate leads to the decrease of crystalinity. Another fact that could be deduced from the XRD patterns is the broadening, FWHM (Full Width Half Maximum), of the main peak of anatase ((004)) which has been increased from 0.1889 to 0.2808 when synthesized in the electrolyte containing sodium carbonate. According to
Fig. 2. Schematic diagram indicates the microstructure of pure and doped synthesized TiO2 nanotubes.
172
M. Alitabar, H. Yoozbashizadeh / Materials Research Bulletin 95 (2017) 169–176
Fig. 3. (1) X-Ray Map analysis of doped TiO2 nanotubes and (2) EDX analysis of a) pure, doped nanotubes with different concentrations of additive b) 0.02 and c) 0.03 M.
using the anatase diffraction peaks. According to the results of Table 2, the parameters (a, c) enhance by increasing the concentration of additive. Actually the change of parameters could be the result of entering larger ions such as sodium in the lattice of TiO2 nanotubes [38]. Furthermore, the shift of the XRD diffraction peaks of doped nanotubes to the lower angles than the pure one exhibits the creation of expansion in the lattice of nanotubes, hence, it could demonstrate the presence of sodium as dopant in the lattice of nanotubes [39,40]. In fact, this strain could be the reason of destroying the wall of nanotubes when the concentration of sodium carbonate is 0.04 M. 3.2. XPS test
Fig. 4. XRD patterns of pure and doped TiO2 nanotubes in organic electrolyte.
Table 2 Effect of concentration of sodium carbonate on the lattice parameter of TiO2 nanotubes. photocatalyst
a (Å)
C (Å)
TiO2 0.01 M Na2CO3-TiO2 0.02 M Na2CO3-TiO2 0.03 M Na2CO3-TiO2
3.7892 3.8036 3.8144 3.8295
9.4803 9.4931 9.5023 9.5112
the Williamson-Hall equation, the mean size of crystallites [25] of TiO2 nanotube was estimated as 92 5 and 77 5 nm for pure and doped samples, respectively. In addition, based on this procedure the strain of structure was calculated as 0.095 and 0.14% for the mentioned samples, respectively. Therefore, by considering these results and FESEM images it can be suggested that using the additive leads to development of some expansion and consequently, strain in the lattice of TiO2 nanotubes [38]. The lattice constant parameters (a,c) for doped nanotubes were calculated
To investigate the chemical state of elements and the surface composition of the synthesized nanotubes, the XPS (X-ray Photoelectron Spectroscopy) analysis was performed. Fig. 5 illustrates the XPS spectra of Ti2p, Na1s, O1s and C1s for doped TiO2 nanotube arrays after annealing at 500 C for 2 hours. In Fig. 5a two peaks could be observed at 458.43 and 464.14 eV which are assigned to Ti2p3/2 and Ti2p1/2, respectively. The distance between them is about 5.71 eV confirming that the main chemical state of Ti in TiO2 nanotube lattice is Ti4+ [41]. A high resolution spectrum of sodium has been illustrated in Fig. 5b. The peak located at 1069.95 eV corresponds to Na1s showing the presence of Na in the lattice of TiO2 nanotubes [42]. Furthermore, according to Fig. 6 the peaks of Ti 2p(3/2) and Ti 2p(1/ 2) for pure nanotubes locate at 459.10 and 464.80 ev, respectively which are in agreement with the literature reported values [43]; these peaks for doped samples shift to the lower values which could be a result of reduction of Ti4+ to Ti3+. In fact, the incorporation of sodium ions in the lattice of nanotubes yields to the creation of oxygen vacancies in the doped samples. Accordingly, the negative charge introduced in the lattice of nanotubes is instead compensated by conversion of Ti4+ to Ti3+ [44]. To confirm this, the high resolution spectrum of oxygen in the doped nanotubes has been investigated as well [5,45]. Fig. 5c shows the high-resolution XPS spectrum of O1s. The O1s peaks locate at 530.50, 531.50 and 533.30 eV which belong to Ti-O bonds (existence of oxygen in crystal lattice of nanotubes), oxygen faults and the hydroxyl groups obtained from chemisorbed water, respectively [46]. The area under the peak situated at 531.5 eV is
M. Alitabar, H. Yoozbashizadeh / Materials Research Bulletin 95 (2017) 169–176
173
Fig. 5. XPS test of doped TiO2 nanotubes. (a) titanium, (b) sodium, (c) oxygen and (d) carbon.
Fig. 6. XPS test of Ti2p1/2, Ti2p3/2 for pure and doped TiO2 nanotubes.
sodium ions inside the lattice of TiO2 nanotubes (not on their surface) could be proved. Fig. 5d illustrates the XPS spectra of TiO2 nanotubes doped with C. As shown in this figure, five peaks, one strong peak at bonding energy of 284.50 eV and four weak peaks at bonding energies of 282.47, 286.47, 287.90 and 288.97 eV could be distinguished. The peak observed around 284.50 eV is assigned to Carbon absorption on the surface as contamination (C C bonds). The peaks at 287.90 and 288.97 eV are attributed to C—O bond (hydroxyl composition) and C¼O bonds (Carboxylic groups), respectively. The observed peaks situating at 286.47 and 282.87 eV are ascribed to Ti-C O as carbonate species and Ti-C bonds, respectively [47]. The carbonate species could be formed as carbon atoms diffuse into the interstitial locations of the TiO2 lattice. However, the presence of C atoms inside the TiO2 nanotubes lattice could not be proved. To confirm the presence of carbon in the lattice of TiO2 nanotube, the existence of a peak in the range of 281.5 283 eV is necessary [5,6]. Here, a very small peak detected at 282.47 eV which is related to TiC bonds in the nanotube lattice indicates the low amount of doped carbon in TiO2 nanotubes. The origin of doped carbon could be related to the ethylenhlycol which exists in the organic electrolyte. It should be mentioned that the results of XRD and FE-SEM analyses also confirmed this fact. 3.3. Uv-visible tests
important parameter, since it is ascribed to the oxygen vacancy [45], which can be developed during the metal and none metal doping process in the lattices of TiO2 nanotubes [47]. Finally, using the results of XPS, XRD and FESEM analyses, the presence of
Fig. 7 exhibits the absorption edge (in the range of 250–690 nm) and the band gap energy of pure and doped nanotubes. It could be seen in Fig.7a that the strong absorption was occurred in the wavelength range of almost uv-light (l < 400 nm) and the weak
174
M. Alitabar, H. Yoozbashizadeh / Materials Research Bulletin 95 (2017) 169–176
Fig. 7. a) uv-visible tests and b) DRS plots of pure and doped TiO2 nanotubes.
absorption was happened in the visible region. It could be said that the strong absorption is due to intrinsic inter-band transition while the weak absorption can be assigned to the forming of faults depending on the synthesis method [48]. As could be observed, in the plots of doped nanotubes, all of samples exhibit stronger absorption in the range of 400–690 nm. Comparing the results of doped nanotubes shows the enhancing of absorption by increasing the additive concentration from 0.02 to 0.03 M. However, among of the doped samples the largest optical absorption edge belongs to the sample synthesized in the electrolyte containing 0.03 M of additive. Furthermore, inserting the additive shifts the absorption edge of pure nanotubes by about 92 nm to the longer wavelength (from 396 to 488 nm). To determine the effect of additive on the band gap energy of TiO2 nanotubes, DRS (Diffuse Reflectance Spectra) analysis was applied which is indicated in Fig. 7b. The band gap energy was calculated using Kubak Munk function and Tauc plot. Based on the Kubak Munk function, a= (1-R)2/(2R), where R is diffuse reflection. In Tauc plot (ahv) 0.5 is drawn versus hv, where h is the Pelank constant. Actually, by extrapolating the linier portion of (ahv) 0.5 graph to the zero value, the band gap energy could be estimated [49]. As Fig. 7b illustrates, the band gap energy of TiO2 nanotubes without any additives is equal to 3.20 eV for anatase phase. Clearly, the use of sodium carbonate reduced the band gap energy of TiO2 nanotubes from 3.20 to 2.88 eV. This reduction could be related to the presence of sodium and carbon in the lattice of TiO2 nanotubes. It seems that the sodium develops some trap levels between valance and conductive bands close to the conduction band. Therefore, it induces the excitation of electrons from the valance band to the trap levels [40,50]. In addition, mixing the 2p state of carbon with 2p state of oxygen creates new valance band upwards the valance band of the anatase phase [51,52].
lower potential is needed to separate the photoinduced electrons and inject them to the outer circuit in doped TiO2 nanotubes [51– 54] Generally, doping process on the TiO2 nanotube arrays leads to the formation of oxygen vacancy and consequently, the creation of unpaired electrons or Ti3+ centers to keep the charge balance. These vacancy states could act as the trapper center of electrons. sodium ions are homogenously dispersed [55], since they could capture the electrons and holes and thus, retard the recombination of electron-hole pairs, thereby, the photocatalytic activity enhances [56]. Furthermore, the substitution of oxygen with carbon atoms in TiO2 lattice leads to the mixture of 2p states of carbon with 2p states of oxygen and development of the new valance band (VB) [16,57], hence, the VB edge of doped TNA moves upwards and the electrons in 2p states of carbon can excited with visible light. In other words, the recombination rate of electron-hole pairs decreases in the doped TiO2 [33]. This implies that more photoelectrons-hole pairs have been generated under the radiation of solar light which could participate in photocatalytic reactions. Overally, the photocatalytic activity of a material means its ability to create electron-hole pairs [50,55]. Actually, the rate of generating of charge carries, which transfer from interior to the surface of photoanode, should be higher than the rate of recombination of photo induced electrons and holes [40,41].
3.4. Effect of additive on the photocurrent density The photocurrent measurement has been utilized in order to determine the convertion ability of photons to current in doped TiO2 nanotubes. Fig. 8 displays the linear sweep voltammograms (LSV) for different samples under visible light irradiation. As it is seen, in the constant applied potentials, the photocurrent density of all doped TiO2 nanotube arrays is significantly higher than that of pure TiO2 nanotube array. The shift of zero current potential to the more negative values in the doped samples comparing to the pure TiO2 nanotubes (from 0.20 V for pure nanotubes to 0.58 V for doped nanotubes) may be related to the combined effect of Na and C atoms in the lattice of doped nanotubes. This means that the
Fig. 8. Photocurrentdensity as a functional of potential for pure and doped TiO2 nanotubes.
M. Alitabar, H. Yoozbashizadeh / Materials Research Bulletin 95 (2017) 169–176
175
photocatalytic activity of nanotubes enhanced. The photocurrent density in doped nanotubes was about 5 times higher than that of pure one indicating the positive effect of sodium Carbonate. Acknowledgment The authors are grateful to the extracting and corrosion laboratories of Sharif University of Technology. Moreover, we would like to give our special thanks to Mr. M. Farokhgisoo and Mrs. Familifard. Appendix A. Supplementary data
Fig. 9. Photocurrent density as a function of time for pure and doped TiO2 nanotubes.
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j. materresbull.2017.07.013. References
According to the results, with doping the TiO2 nanotube arrays with sodium and carbon, the recombination rate of electron-hole pairs reduces and the density of charge carrier existing on the surface of photoanode increases which accelerates the water splitting reactions on its surface [50]. Fig. 9 displays the photocurrent density of pure and doped nanotubes (0.03 M) as a dependence of time in the electrolyte containing KOH and 10%wt ethylenglycol under the visible light irradiation (100 mw/cm2) at 0.5 V. As.Ag/AgCl [49] it is observed that the photocurrent density of doped samples is more than 5 times than of that of pure sample, which shows the positive effect of additive, this is in agreement with photocurrent density-voltage results. Many factors can improve the photocatalytic activity of nanotubes which are synthesized in the organic electrolyte containing sodium carbonate as an additive. According to the results, by irradiation the energy equal or higher than the band gap energy of TiO2 nanotubes, it is possible to generate the electrons and holes in the conduction band (CB) or valance band (VB), respectively. Reducing the band gap energy by doping sodium and carbon could reduce the energy needed for exciting the electrons and holes. Furthermore, sodium and carbon act as trapper center for electron and holes yielding to the reduction of recombination rate and increasing the photocatalytic activity of nanotubes, thus, the excitation range of electron-hole pairs enhances from uv-light to uv-visible range. Obviously, the length and crystallite size of nanotubes is effective parameter on the photocatalytic activity. It has been shown that by increasing the length and decreasing the crystallite size [58] of nanotube the photocatalytic activity enhances. 4. Conclusion In this work, anodizing method was used to synthesize the TiO2 nanotubes in organic electrolyte. When the nanotubes are applied as photoanode in water splitting process, it is necessary to reduce the band gap energy and enhance the photocatalytic activity of nanotubes. In this research, the anodizing time and band gap energy of TiO2 nanotubes reduced simultaneously by using sodium carbonate as an additive. The use of this additive led to development sodium and carbon co-doped nanotube arrays, accordingly, the band gap energy reduced. The results revealed the development of Ti-C, Ti-C O bonds, the existence of a peak at 1072.5 eV confirmed the presence of sodium and Carbon in nanotubes lattice. Furthermore, using this additive increased the separation rate of electrons and holes. Moreover, the
[1] O.K. Varghese, M. Paulose, K. Shankar, G.K. Mor, C.A. Grimes, Nanosci. Nanotechnol. 5 (2005) 1158. [2] Y. Li, Y. Xiang, S. Peng, X. Wang, L. Zhou, Electrochim. Acta 87 (2013) 794–800. [3] R. Beranek, J.M. Macak, M. Gärtner, K. Meyer, P. Schmuki, Electrochim. Acta 54 (2009) 2640–2646. [4] H. Yan, X. Wang, M. Yao, X. Yao, Prog. Nat. Sci. Mater. Int. 23 (2013) 402–407. [5] M. Mollavali, C. Falamaki, S. Rohani, Int. J. Hydrogen Energy 40 (2015) 12239– 12252. [6] M. Mollavali, C. Falamaki, S. Rohani, Int. J. Hydrogen Energy 41 (2016) 5887– 5901. [7] A. Fujishima, K. Honda, Nature 238 (1972) 37–38. [8] Angelis De Filippo, I. Fantacc, Selloni Annabella, K. Mohammad, M. Grätzelx, J. Phys. Chem. C 114 (2010) 6054–6061. [9] Kesong Yang, Ying Dai, Baibiao Huang, S. Han, Phys. Chem. B 110 (2006) 24011– 24014. [10] S. Li, S. Lin, J. Liao, N. Pan, D. Li, J. Li, Int. J. Photoenergy 2012 (2012) 1155. [11] L. Wenjun, Y. Tao, W. Yaoming, L. Zhaosheng, Y. Jinhua, Z. Zhigang, J. Phys. D: Appl. Phys. 40 (2007) 1091. [12] T. Umebayashi, T. Yamaki, H. Itoh, K. Asai, J. Phys. Chem. Solids 63 (2002) 1909– 1920. [13] D. Regonini, C.R. Bowen, A. Jaroenworaluck, R. Stevens, Mater. Sci. Eng. R: Rep. 74 (2013) 377–406. [14] J. Xing, Z. Xia, J. Hu, Y. Zhang, L. Zhong, Corros. Sci. 75 (2013) 212–219. [15] J.P. Xu, J.F. Wang, Y.B. Lin, X.C. Liu, Z.L. Lu, Z.H. Lu, L.Y. Lv, F.M. Zhang, Y.W. Du, J. Phys. D Appli. Phys. 40 (2007) 4757. [16] Xiaobo Chen, C. Burda, J. Am. Chem. Soc. 130 (2008) 5018. [17] S. Karthik, T. Kong Chhay, K.M. Gopal, A.G. Craig, J. Phys. D Appl. Phys. 39 (2006) 2361. [18] Z. Zhao, Q. Liu, J. Phys D Appl Phys 41 (2008). [19] M. Guo, J. Du, Phys. B Condens. Matter. B 407 (2012) 1003–1007. [20] Q. Liu, D. Ding, C. Ning, X. Wang, Int. J. Hydrogen Energy 40 (2015) 2107–2114. [21] H. Liu, G. Liu, X. Shi, Coll. Surf. A 363 (2010) 35–40. [22] A.B. Murphy, Sol. Energy Mater. Sol. Cells 91 (2007) 1326–1337. [23] V.A. Tran, T.T. Truong, T.A.P. Phan, T.N. Nguyen, T.V. Huynh, A. Agresti, S. Pescetelli, T.K. Le, A. Di Carlo, T. Lund, S.-N. Le, P.T. Nguyen, Appl. Surf. Sci. 399 (2017) 515–522. [24] O. Ola, M.M. Maroto-Valer, J. Photochem. Photobiol. C Photochem. Rev. 24 (2015) 16–42. [25] W.H. Hall, G.K. Williamson, J. Sci. Instrum. 29 (1952) 132. [26] T. Gheiratmand, H.R.M. Hosseini, P. Davami, F. Ostadhossein, M. Song, M. Gjoka, Nanoscale 5 (2013) 7520–7527. [27] S. Rani, S.C. Roy, M. Paulose, O.K. Varghese, G.K. Mor, S. Kim, S. Yoriya, T.J. LaTempa, C.A. Grimes, Phys. Chem. Chem. Phys. 12 (2010) 2780–2800. [28] K. Lee, A. Mazare, P. Schmuki, Chem. Rev. 114 (2014) 9385–9454. [29] D.P. Padiyan, D.H. Raja, Energy Procedia 22 (2012) 88–100. [30] J.M. Macak, P. Schmuki, Electrochim. Acta 52 (2006) 1258–1264. [31] Baiqi Shao, Song Yan, Mengmeng Jiao, Wei Lü, Hongpeng You, Inorg. Chem. 55 (2016) 1912–1919. [32] A.S. Povarennykh, Crystal Chemical Classification of Minerals, Springer, Ukrainia, 1972. [33] J. Tian, H. Gao, H. Deng, L. Sun, H. Kong, P. Yang, J. Chu, J. Alloys Compd. 581 (2013) 318–323. [34] J.F. Moulder, F. Stickle, E. Sobol, D. Bomben, Handbook of X-ray Photoelectron Spectroscopy, United State of America, 1992. [35] Y. Liao, W. Que, Q. Jia, Y. He, J. Zhang, P. Zhong, J. Mater. Chem. 22 (2012) 7937– 7944. [36] Yulong Liao, Huaiwu Zhang, Jie Li, Guoliang Yu, Zhiyong Zhong, Feiming Bai, Lijun Jia, Shihong Zhang, P. Zhong, J. Appl. Phys. 115 (2014) 304-. [37] F. Marzpour Shalmani, R. Halladj, S. Askari, Ultrason. Sonochem. 29 (2016) 354–362.
176
M. Alitabar, H. Yoozbashizadeh / Materials Research Bulletin 95 (2017) 169–176
[38] Anderson Janotti, Bharat Jalan, Susanne Stemmer, C.G.V.d. Walle, Appl. Phys. Lett. 100 (2012) (262104-262101-262104-262103). [39] W. Li, J. Yang, J. Zhang, S. Gao, Y. Luo, M. Liu, Mater. Res. Bull. 57 (2014) 177–183. [40] S. Shalini, N. Prabavathy, R. Balasundaraprabhu, T. Satish Kumar, P. Walke, S. Prasanna, D. Velayuthapillai, J. Mater. Sci. Mater. Electron. (2016) 1–9. [41] L.G. Devi, R. Kavitha, Appl. Catal. B Environ. 140–141 (2013) 559–587. [42] T.G. Deepak, D. Subash, G.S. Anjusree, K.R.N. Pai, S.V. Nair, A.S. Nair*, ACS Sustainable Chem. Eng. 2 (2014) 2772–2780. [43] M. Nischk, P. Mazierski, Z. Wei, K. Siuzdak, N.A. Kouame, E. Kowalska, H. Remita, A. Zaleska-Medynska, Appl. Surf. Sci. 387 (2016) 89–102. [44] S. Munir, S.M. Shah, H. Hussain, R. Ali khan, Mater. Des. 92 (2016) 64–72. [45] W. Zhongbiao, D. Fan, Z. Weirong, W. Haiqiang, L. Yue, G. Baohong, Nanotechnology 20 (2009) 235701. [46] N. Yan, Z. Zhu, J. Zhang, Z. Zhao, Q. Liu, Mater. Res. Bull. 47 (2012) 1869–1873. [47] S. Sakthivel, H. Kisch, Angew. Chem. Int. Ed. 42 (2003) 4908–4911. [48] S. Zhang, F. Peng, H. Wang, H. Yu, S. Zhang, J. Yang, H. Zhao, Catal. Commun. 12 (2011) 689–693.
[49] H. Ali, N. Ismail, M. Mekewi, A.C. Hengazy, J. Solid State Electrochem. 19 (2015) 3019–3026. [50] N.L. Huan Xie, Baoshun Liu, Jingjing Yang, Xiujian Zhao, J. Phys. Chem. C 120 (2016) 10390–10399. [51] Y.L. Pang, S. Lim, H.C. Ong, W.T. Chong, Appl. Catal. A Gen. 481 (2014) 127–142. [52] R. Yuan, B. Zhou, D. Hua, C. Shi, Front. Environ. Sci. Eng. 9 (2015) 850–860. [53] R. Asahi, T. Morikawa, T. Ohwaki, K. Aoki, Y. Taga, Science 293 (2001) 269–271. [54] Xiaoping Han, G. Shao*, J. Phys. Chem. C 115 (2011) 8274–8282. [55] M. Zhou, J. Yu, J. Hazard. Mater. 152 (2008) 1229–1236. [56] Zheshuai Lin, Alexander Orlov, Richard M. Lambert, M.C. Payne, J. Phys, Chem. B 109 (2005) 20948–20952. [57] F. Esaka, K. Furuya, J. Vac. Sci. Technol. A Vac. Surf. Films 15 (1998). [58] Mahamasuhaimi Masae, Pichaya Pitsuwan, Chaturong Pholthawon, Niti Pawanwatcharakorn, Lek Sikong, P. Kongsong, Thammasat Int. J. Sci. Technol. 20 (2015) 63–71.