Enhancement of the photocatalytic activity of TiO2 by doping it with calcium ions

Journal of Colloid and Interface Science 357 (2011) 168–178

Contents lists available at ScienceDirect

Journal of Colloid and Interface Science www.elsevier.com/locate/jcis

Enhancement of the photocatalytic activity of TiO2 by doping it with calcium ions U.G. Akpan, B.H. Hameed ⇑ School of Chemical Engineering, Engineering Campus, Universiti Sains Malaysia, 14300 Nibong Tebal, Penang, Malaysia

a r t i c l e

i n f o

Article history: Received 3 September 2010 Accepted 5 January 2011 Available online 21 January 2011 Keywords: Photocatalytic Degradation Cyclic heat treatment Acid red 1 Doping TiO2

a b s t r a c t Titanium dioxide (TiO2) with an enhanced photocatalytic activity was developed by doping it with calcium ions through a sol–gel method. The developed photocatalysts were characterized by Fourier transform infrared (FTIR) spectroscopy, N2 physisorption, X-ray photoelectron spectroscopy (XPS), and X-ray diffraction. Their surface morphologies were studied using surface scanning electron microscopy (SEM). The XPS analyses confirmed the presence of Ti, O, Ca, and C in the Ca-doped TiO2 sample. The activities of the catalysts were evaluated by photocatalytic degradation of an azo dye, acid red 1 (AR1), using UV light irradiation. The results of the investigations revealed that the samples calcined at 300 °C for 3.6 h in a cyclic (2 cycles) mode had the best performance. Lower percentage dopant, 0.3–1.0 wt.%, enhanced the photocatalytic activity of TiO2, with the best at 0.5 wt.% Ca–TiO2. The performance of 0.5 wt.% Ca–TiO2 in the degradation of AR1 was far superior to that of a commercial anatase TiO2 Sigma product CAS No. 1317-70-0. The effect of pH on the degradation of AR1 was studied, and the pH of the dye solution exerted a great influence on the degradation of the dye. Ó 2011 Elsevier Inc. All rights reserved.

1. Introduction The photocatalytic potentiality of titanium dioxide (TiO2) has extensively been studied [1–12]. Research has indicated that the band gap of TiO2 encourages recombination of holes (h+) and electrons (e) generated by incident of photon rays on it, thus limiting its photocatalytic abilities. Photocatalysis as it is known is a photoinduced reaction which is accelerated by the presence of a catalyst [13]. Photocatalytic reactions are activated by absorption of a photon with sufficient energy (equals or higher than the band-gap energy (Ebg) of the catalyst) [14]. The absorption leads to a charge separation due to promotion of an electron (e) from the valence band of the semiconductor catalyst to the conduction band, thus generating a hole (h+) in the valence band. The activated electron will react with an oxidant to produce a reduced product, while the hole is to react with a reductant to produce an oxidized product. Titanium dioxide (TiO2) photocatalyst is chemically and biologically inert, photocatalytically stable, relatively easy to produce and to use, able to efficiently catalyze reactions, cheap, and without risk to the environment or humans. Nevertheless, these advantages do not forestall its photocatalytic limitations by a recombination center existing between the activated electrons (e) and the generated holes (h+) because of its high band gap. But recent literature [15–25] has shown that there is a great

⇑ Corresponding author. E-mail address: [email protected] (B.H. Hameed). 0021-9797/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.jcis.2011.01.014

enhancement in the photocatalytic activity of TiO2 by doping it with either metal or nonmetal ions or both. Furthermore, TiO2 photocatalyst has been employed in several studies which safeguard the environment from the deleterious effects of pollution caused by the release of some liquid effluents into the water channels. Most of these effluents released into waterways by some industries contain dyes and toxic substances which are harmful to the health and general well-being of man. Therefore, enormous demand is placed on the scientists and environmentalists to see to the proper treatment of wastewaters from these industries before releasing them to the environment. Many methods such as traditional physical techniques (adsorption on activated carbon, ultrafiltration, reverse osmosis, coagulation by chemical agents, ion exchange on synthetic adsorbent resins, etc.) have been used for the removal of dye pollutants [26,27]. These methods only succeed in transferring organic compounds from water to another phase, thus creating secondary pollution. This will require a further treatment of solid wastes and regeneration of the adsorbent which will add more cost to the process. Microbiological or enzymatic decomposition [28], biodegradation [29], ozonation [30], and advanced oxidation processes such as Fenton and photo-Fenton catalytic reactions [26,31] and H2O2/UV processes [32] have also been used for removal of dyes from wastewaters. The advantage of photocatalysis in the removal of dyes from wastewaters is embedded in its ability to completely mineralize the target pollutant [33], and TiO2 is considered in this case because of its advantages over the other semiconductor photocatalysts. Many substances ranging from metals to nonmetals and even combinations of the two have already been used in doping

U.G. Akpan, B.H. Hameed / Journal of Colloid and Interface Science 357 (2011) 168–178

Scheme 1. Structure of acid red 1 (AR1) dye.

TiO2 for photocatalytic processes, but very scanty literature is available for the use of calcium ions in doping TiO2. Li et al. [34] reported the use of selected group IIA elements and their performances in doping TiO2. Their effectiveness was evaluated by photocatalytic production of hydrogen and calcium-doped TiO2 was found to strive well. Therefore, considering the associated recombination center problems that exist in TiO2 photocatalyst, this present study was initiated to study the level of enhancement that is required to improve the photocatalytic potentiality of TiO2 by doping it with calcium ions. This will, however, provide a means of treating wastewaters containing dyes, and also test the potentiality of the newly developed calcium titanium dioxide (CaTiO2) photocatalyst on acid red 1 (AR1), an azo dye that was selected for the purpose of this study. The structure of AR1 is given in Scheme 1. 2. Materials and methods 2.1. Preparation of the photocatalysts The method of Li et al. [34] was adopted for the preparation of the photocatalysts. The undoped TiO2 was prepared by the sol–gel method using titanium (IV) butoxide [Ti(OBu)4] (97% reagent grade obtained from Sigma Aldrich Chemicals) as the precursor. Five milliliters of Ti(OBu)4 was dissolved in 20 mL ethanol (95% laboratory R&M chemical, Essek, UK) under stirring and then 0.5 mL HNO3 (Merck, Germany) solution (VHNO3: VH2O = 1:1) was added and stirred for about 5 min. Then, 1 mL double-distilled H2O was added drop by drop to the above solution at a rate of 0.43 mL/min. The mixture was vigorously stirred at room temperature until a trans-

169

parent sol was formed. The gel was obtained after aging the sol for 24 h at room temperature. Thereafter, the gel was dried at 80 °C on a temperature-controlled magnetic stirrer hot plate, ground into fine powder, and calcined in an ISUZU Muffle furnace (Seisakusho Ltd., Tokyo, Japan). The calcium ion-doped TiO2 photocatalysts were prepared in the same way, except that a solution of the calcium chloride CaCl2 dissolved in distilled water was used instead of water. The resulting dried samples were also calcined in the same furnace. The calcinations of the samples were carried out in two modes viz; direct heat treatment and heat treatment in a cyclic mode. This was done to verify the effect of the mode of calcinations on the activity of the catalyst. In the cyclic mode heat treatment, the catalysts were calcined at 300 °C in the furnace programmed for 2 h at a heating rate of 7.5 °C/min and was allowed to cool to 105 °C in 2.9 h. After the first cycle was completed, the second cycle was run for 1.6 h. The direct heat treatment was undertaken by programming the furnace to the required calcination temperature and set calcination time using the same heating and cooling rates as in the cyclic heat treatment, respectively; in this study 300 °C and 3.6 h were the respective calcination temperature and time. In general, the same effective temperature and time were used in each case for easy comparison. It must also be noted that the sample used for these tests was prepared in the same batch, but only divided at the point of calcinations.

2.2. Catalysts characterization Nitrogen adsorption–desorption isotherms of the developed photocatalysts were collected from an ASAP 2020 V3.02 H Micromeritics surface area and porosity analyzer at 77 K. The Brunauer– Emmet–Teller (BET) surface area was calculated from the linear part of the BET plot. The pore-size distribution plots were obtained by using the Barret–Joyner–Halenda (BJH) model. Powder X-ray diffraction (XRD) patterns of the catalysts were measured by D8 Advanced X-ray solution. The elemental compositions of the catalysts were determined using an energy dispersive X-ray (EDX) detector mounted on a microscope. The X-ray photoelectron spectroscopy (XPS) spectra were obtained with a Mesin XPS Omicron els 5000 spectrophotometer using Al Ka at 1480 kV as radiation X-ray source. All binding energies were calibrated to the C 1s peak at 284 eV. The microstructure and morphology of the prepared cat-

Fig. 1. XRD spectra of TiO2 and Ca–TiO2.

170

U.G. Akpan, B.H. Hameed / Journal of Colloid and Interface Science 357 (2011) 168–178

117.8 m2/g maximum reported for a similar process [35]. A single adsorption total pore volume of pores of width less than 1232.66 Å at P/Po = 0.9840 for the 0.5 wt.% Ca–TiO2 was evaluated to be 0.1725 cm3/g, while its Barret–Joyner–Halenda adsorption average pore width is 40.63 Å. The BJH adsorption average pore widths of 30.41 and 32.22 Å were obtained for the pure TiO2 calcined at 300 °C for 2 h in a straight mode and 3.6 h in a cyclic mode, respectively. This offers a bit of insight into the porosity of developed photocatalysts. The prepared photocatalysts were also subjected to X-ray diffraction measurement (using a D8 Advanced X-ray solution) to determine their crystal phase compositions. The results in Fig. 1 show the XRD pattern of various catalysts calcined at 300 °C for 2 h and samples which went through cyclic heat treatment. Only a single anatase phase TiO2 was formed for both pure TiO2 and TiO2-doped with calcium ions; no calcium oxide impurity phase was detected. This observation is in agreement with previous findings [35,36]. It is possible that the dopant could also be intercalated into the TiO2 lattice since the radius of Ca2+ is larger than that of Ti4+ [34]. This was further elucidated by the crystal analysis of the developed photocatalysts (Table 1). All the photocatalysts, whether doped or undoped, had the same tetrahedral crystal structure and a single anatase phase. Nevertheless, the doping of TiO2 with Ca2+ caused some distortion in the lattices of TiO2 and hence, differences in the lattice parameters as noted in Table 1. The doped samples (0.3 wt.% CaTiO2 and 0.5 wt.% CaTiO2) have the same lattice parameters and these are quite different from those of the undoped samples. The FTIR results in Fig. 2 revealed that there are some absorption bands in the regions of 486–541, 1066–1175, 1614–1633, 2345–2376, and 3736–3856 cm1 of the spectra. Previous works [37,38] revealed that the absorption bands in the region of 3420–

alysts were observed using a Philips Model XL30S surface scanning electron microscopy. The active surface functional groups present in the catalysts were determined by the Fourier transform infrared (FTIR). The spectra were recorded in the range of 4000–400 cm1. 2.3. Photocatalytic activities of the catalysts The photocatalytic activities of the photocatalysts were performed at room temperature (27 °C) in a 400-mL jacketed glass reactor fitted with a 9 W 5-in.-long Philips (PL-S 9 W/10/2P Hg, maximum absorption wavelength at 254 nm) bulb made in Poland. Then 300 mL of the required concentration of an unadjusted pH of AR1 was poured into the photoreactor (which was placed in a black box to shield the researcher from direct contact with the UV light) and after the addition of 0.20 g of the catalyst to it, the light was switched on. The reactor content was agitated at 630 rpm using a magnetic stirrer and air was introduced into the reaction medium through an air pump. Samples were withdrawn from the irradiated solution at preset time intervals, filtered with 0.45-lm Whatman PTFE filter and analyzed for the concentration of the AR1 in the solution at 505 nm (the dye maximum absorption wavelength) using a computer software attached to UV–Vis spectrophotometer, UV-1700 PharmaSpec, Shimadzu. 3. Results and discussion 3.1. Characterization of the catalysts The Brunauer–Emmet–Teller surface area for 0.5 wt.% Ca–TiO2 was calculated to be 147.2 m2/g. This shows that it has a very large surface which is available for the photoreaction as against Table 1 Crystal analysis of the developed photocatalysts.

*

Mode of sample calcination

Lattice parameters (Å) a

b

c

Pure TiO2 @ 300 °C for 2 h straight Pure TiO2 @ 300 °C for 3.6 h cyclic 0.3 wt.% CaTiO2 @ 300 °C for 3.6 h cyclic 0.5 wt.% CaTiO2 @ 300 °C for 3.6 h cyclic

3.7852 3.7845 3.7770 3.7770

3.7852 3.7845 3.7770 3.7770

9.5139 9.5143 9.5010 9.5010

Crystal ID number

Wavelength (nm)

Cell volume (Å3)

00–021–1272(⁄) 01–078–2486(C) 01–089–4921(C) 01–089–4921(C)

1.5406 1.5406 1.5406 1.5406

136.31 136.27 135.54 135.54

Un-identified number.

Fig. 2. FTIR spectra of the catalysts.

171

U.G. Akpan, B.H. Hameed / Journal of Colloid and Interface Science 357 (2011) 168–178

3450 and 1630–1640 cm1 are respectively assigned to the stretching and bending vibrations of the hydroxyl on the surface of TiO2 catalysts, and that the absorption bands in the region of 520–580 cm1 are assigned to the stretching vibration of Ti–O. Therefore, the absorption bands in the region of 1614–1633 cm1 are attributed to the bending vibration of the hydroxyl on the surface of TiO2-based catalysts, while the bands in the region of 3736– 3856 cm1 may be assigned to the stretching vibration of the hydroxyl on the surface of the catalysts, since there was no absorption peaks in the region of 3420–3450 cm1. The absorption bands in the region of 486–541 cm1 are assigned to the stretching vibration of Ti–O. One of the most important features was also observed with the FTIR results; that is, the prominent peaks in each of the spectra acted as indicators to their effectiveness in photocatalytic activities. Thus, photoactivity of the catalysts increased with a reduction in the intensity of the peaks at the band region of 1066– 1175 cm1. Pure TiO2 calcined at 300 °C for 2 h had the highest peak and was less effective than others, followed by the pure TiO2 which went through the cyclic heating during calcination. The sample with the least peak is the TiO2-doped with 0.5 wt.% Ca2+, but had the best performance during the photocatalytic degradation of AR1. This claim was further quantified by evaluating the percentage degradation at 90 min irradiation by each of the catalysts per unit area of its peak. The results are shown in Table 2. It is vividly clear from Table 2 that percentage degradation per unit area of peak increases with decrease in height of the peak. Energy dispersive X-ray analysis of the prepared catalysts provides additional justification for the compositions of the fine particles supported on the titanium dioxide [18]. Fig. 3 shows the EDX spectrum of the 0.5 wt.% Ca–TiO2. The results of the analysis re-

vealed that the percentage of calcium in the prepared catalyst was estimated by the EDX to be 0.55 wt.% as against 0.5 wt.% experimental value. All the elements (calcium, titanium, and oxygen) present in the catalyst were in their K shells. The surface scanning electron microscopy of the catalysts was undertaken to study the microstructure and morphology of the catalysts, and the results are given in Fig. 4. Fig. 4a–d are the SEM of pure TiO2 subjected to a 2-h straight heat treatment at 300 °C at a stretch, and pure TiO2, 0.3 wt.% Ca–TiO2, and 0.5 wt.% Ca–TiO2, respectively; all were subjected to cyclic heat treatment. The results revealed that surfaces of the catalysts are made up of a large number of aggregates of catalyst particles agglomerated together. It also revealed the porous nature of the catalyst. It can be seen that the pure TiO2 which was treated at 300 °C for 2 h was less porous than its counterpart that underwent a cyclic heat treatment at the same temperature, but for 3.6 h. Though the cyclic heated pure TiO2 appeared to be more porous, it had clumsy and larger particles, and this may offer an explanation for the reason it performed better than its 2-h straight heated sample and less than the 0.3 wt.% Ca–TiO2 and 0.5 wt.% Ca–TiO2. This actually confirmed the results of the N2 physisorption which shows that the pure TiO2 that was calcined in a cyclic mode had larger pore sizes than the straight calcined sample. A comparison of Fig. 4c and d revealed that 0.3 wt.% Ca–TiO2 is more crystalline and hence less porous than 0.5 wt.% Ca–TiO2. This could be responsible for the better performance of 0.5 wt.% Ca–TiO2 in photocatalytic reaction than the 0.3 wt.% Ca–TiO2 treated under the same conditions. The chemical states and the binding energies of each element in the samples were determined by the use of XPS. Fig. 5 shows the XPS survey spectra of the doped and undoped TiO2. All the

Table 2 Comparison of the% degradation with considered peaks from FTIR spectra. Mode of sample calcination

% Degradation

Area of peak (% transmittance/cm)

% Degradation/peak area  103

Pure TiO2 – 2 h straight Pure TiO2 – 3.6 h cyclic 0.3 wt.% CaTiO2 – 3.6 h cyclic 0.5 wt.% CaTiO2 – 3.6 h cyclic

54.34 82.22 99.78 99.88

20,245 24,750 24,844 20,417

2.684 3.322 4.016 4.892

Fig. 3. EDX of 0.5 wt.% Ca–TiO2 catalyst calcined in cyclic mode.

172

U.G. Akpan, B.H. Hameed / Journal of Colloid and Interface Science 357 (2011) 168–178

Fig. 4. SEM micrographs of different TiO2 samples: (a) pure TiO2 calcined at 300 °C for 2 h; (b) pure TiO2 subjected to cyclic heat treatment at 300 °C for 3.6 h; (c) 0.3 wt.% Ca– TiO2 subjected to cyclic heat treatment at 300 °C for 3.6 h; (d) 0.5 wt.% Ca–TiO2 subjected to cyclic heat treatment at 300 °C for 3.6 h.

elements (Ti, O, and Ca) used in the preparation of the catalysts and an adventitious C were confirmed to exist in the doped sample. Fig. 6 is the high resolution XPS spectra of 0.5 wt.% Ca–TiO2 nanoparticles at (a) Ti 2p, (b) C 1s, (c) O 1s, and (d) Ca 2p core levels. Two prominent peaks are observed at 459.6 and 465.2 eV binding energies for the Ca-doped TiO2 (Fig. 6a), while one prominent peak is observed at 459.3 eV binding energy and two less prominent peaks at 465.6 and 467.3 eV for the undoped sample. The second distinct peak in the doped sample showed that Ca has fully integrated itself into the lattices of the TiO2 and, thus improved the surface structure of the TiO2. The peaks at 459.6 or 459.3 and 465.2 or 465.6 are respectively due to Ti4+ 2p3/2 and Ti4+ 2p1/2 [36]. Peng et al. [39] reported that most references agreed on the lower binding energy of Ti4+ 2p in N–TiO2, in which Ti4+ 2p3/2 and Ti4+ 2p1/2 core levels can decrease by 0.5–2 eV. In order to ascertain the effect of doping TiO2 with Ca, the C 1s level effect was measured and from Fig. 6b two distinct peaks at energy levels of 285.7 and 286.7 eV were found inside the promi-

nent peak which is at 286.5 eV energy level. This observation is not the same for the undoped TiO2 (figure not shown). Though three different peaks are observed, their energy levels are not distant from each other. Also, in both the doped and the undoped samples, a peak is observed at 289.5 and 288.2 eV, respectively. Peaks at these points indicate the presence of C–O bonds [40]. The prominent peak observed at 533.5 eV in Fig. 6c is ascribed to O 1s electron binding energy of TiO2 and Ca in the doped sample. One other peak observed at 530.9 eV is reported to represent the surface-adsorbed O atom [41]. Other researchers [39] attributed this to Ti–O–Ti linkages in TiO2. Fig. 6d shows the high resolution XPS of Ca–TiO2 at the range of 340 to 355 eV. The binding energies existing in Ca–TiO2 are between those of TiO2 and CaTiO3. Also, the binding energies of Ca2+ 2p3/2 are between CaO and CaTiO3. Chen et al. [42] noted these also in the results of their findings on the influence of calcium ion deposition on the apatite-inducing ability of porous titanium for biomedical applications. Therefore, the peak at 348.8 eV is a

U.G. Akpan, B.H. Hameed / Journal of Colloid and Interface Science 357 (2011) 168–178

173

Fig. 5. XPS survey spectra of 0.5 wt.% Ca-doped and undoped.

Fig. 6. High resolution XPS spectrum of 0.5 wt.% Ca–TiO2 nanoparticles—cyclic heat treated: (a) Ti 2p; (b) C 1s; (c) O 1s; and (d) Ca 2p core levels.

representation of Ca2+ 2p3/2 of CaO. At a binding energy of Ca2+ 2p1/ 2 (351.8 eV) another broader and lower intensity peak is observed. This implies that Ca exists in two different modifications. Armitage et al. [43] identified the binding energy at 347.5 as Ca 2p3/2 which is probably largely attributed to CaO (347.1 eV). They also noted that the presence of CaTiO3 (346.9 eV) could not be ruled out.

3.2. Photocatalytic activity of the catalysts Preliminary experiments revealed that 300 °C was the best calcination temperature for the catalysts (pure TiO2 and the Ca2+doped TiO2) prepared in this study. Hence, the prepared catalysts were calcined at 300 °C for 2 h in a cyclic mode (in a furnace

174

U.G. Akpan, B.H. Hameed / Journal of Colloid and Interface Science 357 (2011) 168–178

programmed for 2 h at a heating rate of 7.5 °C/min and was allowed to cool to 105 °C in 2.9 h, and after the first cycle was completed, the second cycle was run for 1.6 h). Fig. 7 shows the comparison between the catalyst calcined at 3.6 h straight and the cyclic heat-treated catalyst using the same calcination time and temperature. It can be deduced from the results that the performance of the catalyst that went through the cyclic heat treatment was better than the one that was heated straight for the same period. At present, no reason could be adduced for this observation most especially as catalysts which went through straight calcination at the same temperature for 3 and 4 h, respectively, did not perform better (results not shown). Nevertheless, it is thought that there existed a kind of synergistic effect of the residual heat in the catalyst after the first 2 h, as the first cycle was completed at about 105 °C, and the second cycle started. Therefore, all other catalysts used in this study were subjected to cyclic heat treatment for the same period.

To fully study the photocatalytic potentialities of the catalysts, at each photocatalytic experiment, 0.20 g of the catalyst was added to 300 mL of 23.4 mg/L AR1 and subjected to UV light irradiation at room temperature (27 °C). Samples were drawn from the reactor at preset time intervals and the amount of AR1 in the reaction solution was measured by a UV–Vis spectrophotometer 1700 series. In this way the activity of each catalyst was measured and the results were recorded (Fig. 8). Photolysis and dark experiments were conducted in order to fully establish that the reaction was truly photocatalytic and not just loss of color due to only light irradiation and/or due to ordinary adsorption. It can be seen from the results (Fig. 8) that there was no loss of color when the dye solution without catalyst was irradiated for 90 min. Also, the catalytic threshold of the 0.5 wt.% Ca–TiO2 stood at less than 3% dye removal. From Fig. 8, it is clear that there was a little adsorption in the first 10 min of the dark experiment, thereafter desorption set-in, and adsorption–desorption

Fig. 7. Comparison between cyclic and noncyclic heat-treated catalysts.

Fig. 8. Effects of dopant concentrations on the activity of the catalysts in the degradation of AR1.

U.G. Akpan, B.H. Hameed / Journal of Colloid and Interface Science 357 (2011) 168–178

equilibrium was established between 20 and 30 min of the process. The results (Fig. 8) also revealed that the photocatalytic activity of TiO2 in the degradation of AR1 was greatly enhanced by doping it with Ca2+ ions at wt.% as low as 0.3–1.0 wt.%, any further increase to 1.5 and 2.0 wt.% became detrimental to its activity. However, lower percentage dopant showed an increased photocatalytic activity, with 0.3 and 0.5 wt.% Ca–TiO2 proved to be the best. From Fig. 8, it is clearly seen that there was a shift and an increase in the rate of degradation of the AR1 by 0.3 and 0.5 wt.% Ca–TiO2 as there was a total degradation in 90 min UV light irradiation as against 140 min irradiation for the undoped (pure) TiO2. Since the 0.3 and 0.5 wt.% Ca–TiO2 proved to be the best catalysts which accomplished a complete degradation of AR1 at 90 min UV light irradiation, they were further subjected to another test in order to differentiate and choose the best for further studies. Therefore, they were individually employed in the degradation of AR1 of higher initial concentration (50 mg/L).

175

The results as shown in Fig. 9 validated 0.5 wt.% Ca–TiO2 as the best photocatalyst as it accomplished a total degradation of 50 mg/ L of AR1 in 220 min as against 250 min by 0.3 wt.% Ca–TiO2. Previous studies [7,27,34] also showed that their developed photocatalysts (though different from the present study) had the best performance at 0.5 wt.% doping for the degradation of their respective degraded substances. Other studies; such as initial dye (AR1) concentrations and pH effects on the photocatalytic activity of the catalyst were then conducted with 0.5 wt.% Ca–TiO2. The effectiveness of the developed photocatalyst (0.5 wt.% Ca– TiO2) was compared with that of an anatase TiO2 Sigma product CAS No. 1317-70-0. The results (Fig. 10) show that the developed catalysts have better photocatalytic performance than the anatase TiO2 (Sigma product). The results revealed that the commercial TiO2 has a better initial performance in the first 30 min of the process, and thereafter started to diminish and could not attain a complete mineralization of the AR1 dye even after 240 min irradiation.

Fig. 9. Comparison of the activities of 0.3 wt.% Ca–TiO2 and 0.5 wt.% Ca–TiO2 in the degradation of 50 mg/L AR1.

Fig. 10. Photocatalytic activities of 0.5 wt.% Ca–TiO2 and commercial anatase TiO2 Sigma product.

176

U.G. Akpan, B.H. Hameed / Journal of Colloid and Interface Science 357 (2011) 168–178

On the other hand, the 0.5 wt.% Ca–TiO2 proved to be superior to the commercial TiO2 in photocatalytic performance as it attained a complete mineralization of the AR1 dye at 90 min irradiation. To fully investigate the factors which are responsible for the effectiveness of the 0.5 wt.% Ca–TiO2, a pseudo-first-order rate kinetic is necessary. Herrmann [44] recommends that in photocatalysis and other catalytic reactions, rate constant must be evaluated and comparisons made based on the surfaces (active sites) available for the reaction. Hence an integrated pseudo-first-order rate equation (1) was employed to evaluate the pseudo-first-order rate constant.

ln C o =C ¼ kt;

ð1Þ

where C and Co are concentrations in milligrams per liter of dye solution at time t and t = 0, respectively, and k, pseudo-first-order rate constant. A plot of ln{Co/C} against t (Fig. 11) yields a straight line with a slope k. Table 3 shows the pseudo-first-order rate constants, k, BET surface area of catalysts, and the normalized rate constant k0 defined as k0 = k/(BET surface area) for the degradation of AR1 by different catalysts used in this study. It can be seen from the results that doping TiO2 with 0.5 wt.% Ca brought about the increase in its surface area, and hence more active sites are available for the photocatalytic reaction. The Ca-doped TiO2 still proved to be superior to others even after it was normalized per surface area. 3.3. Effects of initial concentration of AR 1 The effect of initial concentrations on the photodegradation of AR1 was studied at 10, 15, 23.4, 30, 40, and 50 mg/L initial concentrations using 0.5 wt.% Ca–TiO2. As can be seen from Fig. 12, there appeared to be an induction time of 10 min for the photocatalytic reaction to fully start up; this may be the reason why the amount of dye degraded (defined as amount degraded = Co–Ct, where

Co = initial amount in mg/L of AR1 and Ct = amount in mg/L of AR1 remaining in solution) in the first 10 min irradiation does not fall into the straight line which characterizes all other points, except for the termination period. This trend reveals that the rate of the degradation of the dye is closely the same, and that one rate expression can describe the whole reaction process. The results in Fig. 12 show that the amount of AR1 degraded is proportional to the irradiation time. The amount degraded for the first 10 min is very high as compared to the amount degraded at the next 10 min and so on. This observation cuts across all initial concentration points considered, and could be a true representation of the whole process. The absorbance spectrum of the AR1 was measured by a UV–Vis spectrophotometer, UV–1700 PharmaSpec, Shimadzu, to fully elucidate the claims of a complete mineralization of the dye by photocatalysis, and the result is shown in Fig. 13. At the end of 180 min UV light irradiation on 40 mg/L AR1, there was no trace of the dye found in the reaction solution as the spectrum at this point was a straight line on the zero absorbance line, regardless of the wavelength. The mineralization of the dye after 90 min irradiation was further verified by total organic carbon (TOC) measurement using TOC–VCSH, Shimadzu analyzer. The TOC values of the dye solution before and after irradiation were 4.224 mg/L and 2.085 mg/L respectively. This shows that more than 50% of the total organic carbon present in the dye solution was removed via the 0.5 wt.% CaTiO2 photocatalytic process. 3.4. Effect of pH on the photocatalytic degradation of AR 1 by 0.5 wt.% Ca–TiO2 A holistic review of the literature revealed that the pH of the solution to be degraded always exerts some influence on the photocatalytic activity of the catalyst [45]. Keeping all other

Fig. 11. Pseudo-first-order kinetic rate plot for various catalysts.

Table 3 Normalized kinetic rate constant based on the BET surface area for catalysts. Catalyst and calcination type

Rate constant (k/min)

BET surface area (m2/g)

Normalized rate constant, g/{(m2) (min)}

Correlation coefficiency (R2)

Pure TiO2 @ 300 °C for 2 h straight Pure TiO2 @ 300 °C for 3.6 h cyclic 0.5 wt.% CaTiO2 @ 300 °C for 3.6 h cyclic Commercial TiO2 (Sigma product)

0.0097 0.0126 0.0280 0.0178

116.34 90.45 147.15 –

8.3E5 13.9E5 19.0E5 –

0.9837 0.9750 0.9644 0.9917

U.G. Akpan, B.H. Hameed / Journal of Colloid and Interface Science 357 (2011) 168–178

177

Fig. 12. Amount of AR1 in mg/L degraded per unit time.

Fig. 13. Absorbance spectra for the degradation of 40 mg/L of AR1 by 0.5 wt.% Ca– TiO2.

parameters (concentration of AR1 and weight of catalyst) constant, the pH of the dye solution was adjusted with either HCl or NaOH to the desired value. The amount of 300 mL of the dye of initial concentration of about 23.4 mg/L was measured and poured into the photocatalytic reactor; hence a fixed weight (0.20 g) of the 0.5 wt.% Ca–TiO2 was added to it and illuminated with UV light. The experiments were performed at initial pH values of 3.0, 5.76, 7.0, and 10.0. The results (Fig. 14) revealed a dramatic degradation at an initial pH value of 3.0 where the percentage degradation (defined as % degradation = (initial concentration–concentration at time t)/initial concentration  100) was 88.0% in 10 min irradiation, whereas only 21.0% degradation was achieved with the unadjusted solution of pH 5.76 at the same irradiation time. At higher initial pH values, 7.0 and 10.0, photodegradation of AR1 was more difficult. The results obtained in this study are in agreement with the findings of other researchers. Sleiman et al. [29] reported a twofold increase in adsorption at pH 4.0 as compared to that at neutral pH. Konstantinou and Albanis [26] reported that the degradation rate of azo dyes increases with decrease in pH. Baran et al. [46] also reported sixfold increase in adsorption efficacy as Bromocresol purple solution was acidified from pH 8.0 to pH 4.5.

Fig. 14. Percentage degradation of AR 1 by 0.5 wt.% Ca–TiO2 (cyclic heat treated) at different solution pH.

178

U.G. Akpan, B.H. Hameed / Journal of Colloid and Interface Science 357 (2011) 168–178

It is established that a strong adsorption of the dye on the TiO2 particles exists at pH < 6 as a result of the electrostatic attraction of the positively charged TiO2 with the dye [26]. At pH > 6.8 as dye molecules are negatively charged in alkaline media, their adsorption is also expected to be affected by an increase in the density of TiO groups on the semiconductor surface. Thus, due to Coulombic repulsion the dye molecules are scarcely adsorbed [47,48], and their degradation becomes difficult. This offers an explanation for why only 52.1% degradation was achieved in 240 min irradiation when the dye solution’s pH was 10, while a complete degradation was achieved at 90 min irradiation when the pH of the dye solution was either 3.0 or 5.76 in this work. 4. Conclusion Titanium dioxide with an enhanced activity has been developed by doping it with 0.5 wt.% calcium. The XRD analysis of the prepared catalysts revealed that only the anatase phase TiO2 was present in all the photocatalysts, whether pure or doped. There was no calcium oxide impurities found in the doped catalysts and this is an indication that calcium ions were integrated into the matrices of the TiO2 during the doping process. The XPS analyses confirmed the presence of Ti, O, Ca, and C in the Ca-doped TiO2 sample. The high resolution XPS also revealed that the doped sample was in its stable state as the presence of Ti4+ and Ca2+ was identified by their binding energies without any interference. The results of the FTIR analysis of the samples offered explanations for the activity of the photocatalysts. The BET surface area for the 0.5 wt.% Ca– TiO2 of 147.2 m2/g shows that enough surfaces are available for the photoreaction. The results obtained in this work proved that the catalysts that went through cyclic calcinations (two cycles) at 300 °C for a total calcination time of 3.6 h are better than the one that was subjected to the same calcination time and temperature in a straight run. A 0.5 wt.% Ca–TiO2 gave the best results in the degradation of AR1. The best results were obtained at lower pH, but neutral and higher pH values were detrimental to the activity of the catalyst since it took a longer time for any reasonable degradation to take place under these conditions. The kinetic rate constants, with or without being normalized, proved that 0.5 wt.% Ca–TiO2 is better than the commercial anatase TiO2 Sigma product CAS No. 1317–70–0 and other developed catalysts in the degradation of AR1 dye, and this is largely due to its large surface area. Acknowledgment The authors acknowledge the research grant provided by Universiti Sains Malaysia under the RU Grant Scheme (RU Grant 814005) which was a great support for this investigation. References [1] V.K. Gupta, R. Jain, A. Mittal, M. Mathur, S. Sikarwar, J. Colloid Interface Sci. 309 (2) (2007) 464–469. [2] G. Li, F. Liu, Z. Zhang, J. Alloys Compd. 493 (2010) L1–L7.

[3] G. Li, L. Lv, H. Fan, J. Ma, Y. Li, Y. Wan, X.S. Zhao, J. Colloid Interface Sci. 348 (2) (2010) 342–347. [4] Z. Ambrus, K. Mogyorósi, A. Szalai, T. Alapi, K. Demeter, A. Dombi, P. Sipos, Appl. Catal. A 340 (2) (2008) 153–161. [5] C.G. Silva, J.L. Faria, J. Mol. Catal. A: Chem. 305 (2009) 147–154. [6] D.L. Liao, C.A. Badour, B.Q. Liao, J. Photochem. Photobiol. A: Chem. 194 (2008) 11–19. [7] A.D. Syafei, C.-F. Lin, C.-H. Wu, J. Colloid Interface Sci. 323 (1) (2008) 112–119. [8] J. Ananpattarachai, P. Kajitvichyanukul, S. Seraphin, J. Hazard. Mater. 168 (2009) 253–261. [9] T.K. Ghorai, D. Dhak, S.K. Biswas, S. Dalai, P. Pramanik, J. Mol. Catal. A: Chem. 273 (2007) 224–229. [10] J.S. Im, S.-M. Yun, Y.-S. Lee, J. Colloid Interface Sci. 336 (1) (2009) 183–188. [11] G. Liu, X. Zhang, Y. Xu, X. Niu, L. Zheng, X. Ding, Chemosphere 59 (2005) 1367– 1371. [12] N.M. Mahmoodi, M. Arami, N.Y. Limaee, N.S. Tabrizi, J. Colloid Interface Sci. 295 (1) (2006) 159–164. [13] A. Mills, L. Hunte, J. Photochem. Photobiol. A: Chem. 108 (1997) 1–35. [14] O. Carp, C.L. Huisman, A. Reller, Prog. Solid State Chem. 32 (1–2) (2004) 33– 177. [15] B. Qi, L. Wu, Y. Zhang, Q. Zeng, J. Zhi, J. Colloid Interface Sci. 345 (2) (2010) 181–186. [16] L.-C. Chen, C.-M. Huang, F.-R. Tsai, J. Mol. Catal. A: Chem. 265 (2007) 133–140. [17] J. Wang, L. Jing, J. Hazard. Mater. 160 (2008) 208–212. [18] G.N. Kryukova, G.A. Zenkovets, A.A. Shutilov, M. Wilde, K. Guther, D. Fassler, K. Richter, Appl. Catal. B 71 (2007) 169–176. [19] G. Liu, X. Wang, Z. Chen, H.-M. Cheng, G.Q. (Max) Lu, J. Colloid Interface Sci. 329 (2) (2009) 331–338. [20] S. Rengaraj, X.Z. Li, P.A. Tanner, Z.F. Pan, G.K.H. Pang, J. Mol. Catal. A: Chem. 247 (2006) 36–43. [21] A. Kumar, N. Mathur, J. Colloid Interface Sci. 300 (1) (2006) 244–252. [22] C.-H. Wei, X.-H. Tang, J.-R. Liang, S. Tan, J. Environ. Sci. 19 (2007) 90–96. [23] H. Yu, X. Zheng, Z. Yin, F. Tao, B. Fang, Chin. J. Chem. Eng. 15 (6) (2007) 802– 807. [24] H. Li, G. Zhao, Z. Chen, G. Han, B. Song, J. Colloid Interface Sci. 344 (2) (2010) 247–250. [25] M. Bettinelli, V. Dallacasa, D. Falcomer, P. Fornasiero, V. Gombac, T. Montini, L. Romanò, A. Speghini, J. Hazard. Mater. 146 (2007) 529–534. [26] I.K. Konstantinou, T.A. Albanis, Appl. Catal. B 49 (2004) 1–14. [27] W.Z. Tang, H. An, Chemosphere 31 (1995) 4158–4170. [28] O.J. Hao, H. Kim, P.C. Chiang, Environ. Sci. Technol. 30 (2000) 449–502. [29] M. Sleiman, D. Vildozo, C. Ferronato, J.-M. Chovelon, Appl. Catal. B 77 (2007) 1– 11. [30] Y.M. Slokar, A.M.L. Marechal, Dyes Pigm. 37 (4) (1998) 335–356. [31] W.G. Kuo, Water Res. 26 (1992) 881–886. [32] N.H. Ince, D.T. Gonenc, Environ. Technol. 18 (1997) 179–185. [33] J. Madhavan, P. Maruthamuthu, S. Murugesan, S. Anandan, Appl. Catal. B 83 (2008) 8–14. [34] Y. Li, S. Peng, F. Jiang, G. Lu, S. Li, J. Serb. Chem. Soc. 72 (4) (2007) 393–402 (JSCS-3569). [35] X. Fan, X. Chen, S. Zhu, Z. Li, T. Yu, J. Ye, Z. Zou, J. Mol. Catal. A: Chem. 284 (2008) 155–160. [36] J. Zhu, Z. Deng, F. Chen, J. Zhang, H. Chen, M. Anpo, J. Huang, L. Zhang, Appl. Catal. B. 62 (2006) 329–335. [37] Y. Li, C. Xie, S. Peng, G. Lu, S. Li, J. Mol. Catal. A: Chem. 282 (2008) 117–123. [38] S. Liu, X. Chen, Chin. J. Catal. 27 (8) (2006) 697–702. [39] F. Peng, L. Cai, L. Huang, H. Yu, H. Wang, J. Phys. Chem. Solids 69 (7) (2008) 1657–1664. [40] W. Ren, Z. Ai, F. Jia, L. Zhang, X. Fan, Z. Zou, Appl. Catal. B 69 (2007) 138–144. [41] C. Liu, X. Tanga, C. Mo, Z. Qiang, J. Solid State Chem. 181 (4) (2008) 913–919. [42] X.-B. Chen, Y.-C. Li, J.D. Plessis, P.D. Hodgson, C. Wen, Acta Biomater. 5 (2009) 1808–1820. [43] D.A. Armitage, R. Mihoc, T.J. Tate, D.S. McPhail, R. Chater, J.A. Hobkirk, L. Shinawi, F.H. Jones, Appl. Surf. Sci. 253 (2007) 4085–4093. [44] J.-M. Herrmann, Appl. Catal. B 99 (2010) 461–468. [45] U.G. Akpan, B.H. Hameed, J. Hazard. Mater. 170 (2009) 520–529. [46] W. Baran, A. Makowski, W. Wardas, Dyes Pigm. 76 (2008) 226–230. [47] M. Stylidi, D.I. Kondarides, X.E. Verykios, Appl. Catal. B 40 (2003) 271–286. [48] Q. Xiao, J. Zhang, C. Xiao, X. Tan, Mater. Sci. Eng. B 142 (2007) 121–125.