Titanium tetraisopropoxide (97%, Sigma-Aldrich, India), Gum Acacia (Finar Chemicals, Ahmedabad, India), isopropanol (99%), glucose, ferric nitrate, sodium hydroxide, ethanol, benzoquinone, ammonium oxalate, and terephthalic acid were obtained from S. D. Fine Chemicals, Mumbai, India. Methyl orange was purchased from Merck, India. Double-distilled water (DDW) was used throughout this experiment. All the reagents were analytical grade and used without further purification.

C-TiO₂ nanoparticles were prepared by sol-gel method. First, 0.1 g of GA was dissolved in 100 mL of DDW (as carbon source), and then, 5 mL of TTIP (as titanium source) was added drop wise to the GA solution under vigorous stirring, which was continuously stirred for 4 h. After that, the formed gel was filtered, washed with DDW several times, and dried in an oven at 120° C for 2 h. The dried sample was then ground and calcined in a muffle furnace at 400° C for 4 h.

FeTiO₃ was synthesized by using the sol-gel method. For the synthesis, 10 mL of 1 mol/L ferric nitrate solution (dissolved in DDW) was added drop wise to 10 mL of 1 mol/L TTIP solution (dissolved in ethanol) and stirred for 3 h. NaOH (1 mol/L) solution was added slowly till precipitation starts, the resulting dark brown colored precipitate was washed with DDW and dried in an oven for 4 h at 120° C. Then, the dried sample was ground and calcined in a muffle furnace at 400° C for 4 h.

Stoichiometric amounts of C-TiO₂ and FeTiO₃ were dispersed in 20 mL of ethanol at room temperature followed by sonication for 1 h to break the agglomerated particles. The mixer was finally stirred at room temperature for 6 h and dried in an oven, resulting in a dry powder after total evaporation of the solvent and calcined at 400° C for 3 h in air atmosphere. In this way, different composites were prepared by varying the amount of FeTiO₃ (10, 20, and 40 wt% FeTiO₃) and denoted as CTFTO-10, CTFTO-20, and CTFTO-40, respectively.

CQD were synthesized according to the literature [31], followed by freeze drying to get powder. In a typical synthesis, a suitable amount of glucose was dissolved in DDW (50 mL) to form a clear solution (1 mol/L). NaOH (50 mL, 1 mol/L) solution was added to the glucose solution, and the mixed solution was treated by an ultrasonic wave for 1 h; then, the sample was adjusted to pH 7 with HCl. After that, the crude solution was filtered with 0.22-µ m filter membrane to remove larger particles and further purified through a dialysis membrane (MWCO 3500) for 1 day. After filter treatment, the obtained solution was freeze-dried in a lyophilizer.

0.05 g CQD was dissolved in 50 mL water to adjust the concentration of CQD solution to 1 mg/mL. CQD/CTFTO-10 composite was synthesized by a simple route in which 0.2 g of CTFTO-10 composite was dispersed in 20 mL CQD solution (1 mg/mL), and the suspension was stirred for 6 h at room temperature and dried in an oven at 80° C for 6 h.

The optical properties of the synthesized catalysts were measured by using UV-Vis DRS spectrophotometer (Shimadzu, UV-3600) with BaSO₄ as the background. Fourier transform infrared spectra of the samples were measured on an IR Prestige-21 spectrometer (Shimadzu). For this, the samples were formed into pellets with KBr. X-ray diffraction pattern was recorded on an X’ pert Pro powder X-ray diffractometer (the Netherlands) using the CuK_α radiation (λ = 0.15406 nm). The morphology, particle size, and elemental composition of the synthesized nanoparticles were characterized by using a transmission electron microscopy (JEOL 2000 FX-II) and scanning electron microscopy (ZEISS, EVO 18) equipped with an energy-dispersive X-ray spectroscopy. The photoluminescence (PL) spectra were measured on a Shimadzu RF-5301PC spectrofluorophotometer, using a Xe lamp as the excitation source at room temperature.

The photocatalytic activity of all the catalysts was tested by the degradation of MO under sunlight irradiation. 0.05 g of catalyst was suspended in 50 mL of an aqueous solution of MO (10 mg/L). Prior to irradiation, the suspension was magnetically stirred in the dark for 1 h in order to reach an adsorption-desorption equilibrium of organic dye on the catalyst surface, and the zero time reading was taken after the equilibrium reached. After that, the suspension was subjected to sunlight irradiation. During irradiation, 3 mL of aliquot samples was withdrawn from the reaction mixture at regular time intervals of 30 min, centrifuged to remove the photocatalyst, and the residual dye concentration was analyzed by UV-Vis spectrophotometer (Shimadzu, UV-2600). All the experiments were performed during the sunny days in the month of June 2015 at Osmania University. The sunlight intensity was measured every hour using an LT Lutron LX-101A digital light meter, and the average light intensity was found to be around 1, 10, 000 lx.

The stability of the CTFTO-10 photocatalyst under sunlight irradiation was investigated by recycling experiments. After the photodegradation, the photocatalyst was recovered by centrifugation, thoroughly washed with DDW, and dried in an oven; then, the dry powder was used in the photocatalytic reaction with a fresh dye solution under the same conditions.

In order to detect the active species generated during the photocatalytic reaction, isopropanol (IPA, 5 mL, ^· OH scavenger), benzoquinone (BQ, 5 mL of 2 mmol/L, ^· O₂^- scavenger), and ammonium oxalate [AO, 0.2 g, hole (h⁺) scavenger] were added to the MO solution [32]. The method was similar to the above photocatalytic activity test. Furthermore, photoluminescence (PL) spectra with terephthalic acid (TA) as a probe molecule were used to disclose the formation of^· OH on the surface of photocatalyst. In a brief experimental procedure, 0.05 g CTFTO-10 was dispersed in 50 mL of the TA (5 × 10^-4 mol/L) aqueous solution with NaOH (2 × 10^-3 mol/L) at room temperature. Prior to irradiation, the suspension was magnetically stirred in the dark. Then, the suspension was subjected to sunlight irradiation, and the PL intensity was measured using a spectrofluorophotometer at an excitation wavelength of 315 nm [33].

Figure 1 shows the XRD patterns of undoped TiO₂, C-TiO₂, FeTiO₃, and C-TiO₂/FeTiO₃ composites. As shown in Fig.1, the XRD patterns of undoped TiO₂ (Fig.1a) and C-TiO₂ (Fig.1b) show the presence of anatase phase of TiO₂ with diffraction peaks at 25.4, 37.8, 48.1, 54.7, 62.8, 70, and 75.3 corresponding the lattice planes (101), (004), (200), (211), (204), (220), and (215), respectively [34]. The diffraction peaks of C-TiO₂ were broader with low intensity compared to undoped TiO₂, indicating the decrease in crystallite size due to the incorporation of carbon atoms into the TiO₂ lattice [35]. Figure 1f indicates the presence of FeTiO₃ phase, with diffraction peaks at 24.9, 32.68, 35.47, and 48.61 corresponding the lattice planes (012), (104), (110), and (024), respectively [27]. The XRD pattern of CTFTO-10 composite obviously did not show any diffraction peaks of FeTiO₃ (Fig.1c), which may be because of low amount of FeTiO₃ and uniform dispersion on the surface of C-TiO₂ [27]. However, the diffraction peaks of FeTiO₃ are visible for CTFTO-20 and CTFTO-40, as a result of high amount of FeTiO₃ (Fig.1d, e). It can also be seen that the diffraction peaks of FeTiO₃ gradually increased with increasing FeTiO₃ amount.

Fig.1

	Figure Option ViewDownloadNew Window
	Fig.1 XRD patterns of undoped TiO2 (a), C-TiO2 (b), C-TiO2/FeTiO3 composites (c-e) and FeTiO3 (f)

Figure 2 shows the diffuse reflectance UV-Vis spectra of undoped TiO₂, C-TiO₂, FeTiO₃, and C-TiO₂/FeTiO₃ composites. The absorption range of C-TiO₂ is located in the visible region, whereas FeTiO₃ possesses strong absorption in the whole solar spectrum. After combining the two semiconductors, the absorptions of C-TiO₂/FeTiO₃ composites within the visible region remarkably increased as compared to C-TiO₂. The absorption intensity of C-TiO₂/FeTiO₃ composites also increased with increasing FeTiO₃ amount (Fig.2a), which subsequently results in the efficient utilization of visible light for the photocatalytic reaction [28]. The band gap energies were calculated by using the formula Ahν = (hν - E_g)^1/2, where A is absorbance, hν is photon energy, and E_g is band gap [36]. Extrapolation of the liner region to zero absorption (Ahν = 0) provides an estimation of band gap energy. The extrapolating graphs are shown in Fig.2b, and the calculated band gap energies were found to be 3.17, 2.824, 2.8, 2.77, and 2.74 for undoped TiO₂, C-TiO₂, CTFTO-10, CTFTO-20, and CTFTO-40, respectively. In comparison with the undoped-TiO₂, the C-TiO₂ showed a redshift in absorption edge, indicating that the carbon doping reduces the band gap of TiO₂ (3.17 eV for undoped TiO₂) by modifying the valance band [37, 38]. The successful incorporation of carbon into the TiO₂ lattice is supported by this observation.

Fig.2

	Figure Option ViewDownloadNew Window
	Fig.2 Diffuse reflectance UV-Vis spectra a and (Ahν )1/2 versus hν plots of undoped TiO2, FeTiO3, C-TiO2, and C-TiO2/FeTiO3 composites b

Figure 3 shows the FTIR spectra of C-TiO₂, FeTiO₃, and C-TiO₂/FeTiO₃ composites. The broad peaks around 3400-3200 cm^-1 and sharp peaks at 1630 cm^-1 are associated with the stretching and bending vibration of surface-adsorbed water and hydroxyl groups of all the samples [37]. The surface hydroxyl groups play an important role in the photocatalytic process, as these groups can be trapped by the holes generated under sunlight irradiation to form hydroxyl radicals which can suppress electron-hole recombination, hence increase the photocatalytic efficiency [37]. The peaks in the region from 400 to 800 cm^-1 of C-TiO₂ are corresponding the asymmetric, symmetric stretching, and the bending mode of Ti-O-Ti bond [39]. In the FTIR spectrum of FeTiO₃, a wide band from 400 to 650 cm^-1 was observed that corresponds to the metal-oxygen bond [40, 41]. In addition, a wide band ranging from 400 to 800 cm^-1 was also observed in the FTIR spectra of all the C-TiO₂/FeTiO₃ composites, which can be assigned to the overlapping peak between the Ti-O-Ti bond and Fe-O bonds, indicating the coexistence of FeTiO₃ and C-TiO₂ [42].

Fig.3

	Figure Option ViewDownloadNew Window
	Fig.3 FTIR spectra of FeTiO3, C-TiO2, and C-TiO2/FeTiO3 composites

Photoluminescence emission spectra can be used to investigate the migration, transfer, and recombination processes of the photogenerated electron-hole pairs in the semiconductors. The PL emission is obtained as a result of recombination of excited electron-hole pair, and hence, lower PL intensity indicates efficient charge carrier separation [37, 43]. Figure 4 shows the PL spectra of undoped TiO₂, C-TiO₂, and C-TiO₂/FeTiO₃ composites. As shown in Fig.4, undoped TiO₂ has a strong emission peak. However, after carbon doping, the PL intensity was decreased, indicating that the carbon doping can effectively inhibit the recombination of photogenerated electron-hole pairs in C-TiO₂ [16]. After the formation of the composites between C-TiO₂ and FeTiO₃, the PL intensities were greatly decreased, suggesting that the recombination rate of the photogenerated electron-hole pairs in C-TiO₂/FeTiO₃ composites is suppressed. This quenching occurs owing to the charge transfer between C-TiO₂ and FeTiO₃. Similar results were also observed in previous report [17]. Therefore, the sunlight-driven photocatalytic activity of C-TiO₂/FeTiO₃ photocatalyst should be higher than that of C-TiO₂.

Fig.4

	Figure Option ViewDownloadNew Window
	Fig.4 PL spectra of undoped TiO2, C-TiO2, and C-TiO2/FeTiO3 composites

Figure 5 shows the morphologies of FeTiO₃, C-TiO₂, and CTFTO-10 obtained by SEM and TEM. In the SEM image of the FeTiO₃ (Fig.5a), irregular agglomerated particles were observed. Figure 5b shows the TEM image of the C-TiO₂, and the particles were spherical in shape, individual as well as a very few number of aggregates with a size of around 10-15 nm, whereas in CTFTO-10 (Fig.5c), few FeTiO₃ particles were uniformly distributed on the C-TiO₂ surface, which indicates the formation of heterojunction between FeTiO₃ and C-TiO₂ particles. In addition to this, EDS was used to determine the chemical compositions of the synthesized samples. According to the recorded EDS pattern, FeTiO₃ consists of iron, titanium, and oxygen elements only (Fig.6a), and C-TiO₂ contains only carbon, titanium, and oxygen elements (Fig.6b), whereas CTFTO-10 contains iron, titanium, oxygen, and carbon elements (Fig.6c), which may confirm the formation of C-TiO₂/FeTiO₃ heterojunction.

Fig.5

	Figure Option ViewDownloadNew Window
	Fig.5 SEM image of FeTiO3 a, TEM images of C-TiO2 b, and C-TiO2/FeTiO3-10 composite c

Fig.6

	Figure Option ViewDownloadNew Window
	Fig.6 EDS of FeTiO3 a, C-TiO2 b, and CTFTO-10 composite c

Figure 7a shows the photocatalytic degradation of MO over undoped TiO₂, C-TiO₂, FeTiO₃, and C-TiO₂/FeTiO₃ composites under sunlight irradiation. As shown in Fig.7a, no obvious degradation was observed when FeTiO₃ was used as a photocatalyst, which can be ascribed to its rapid recombination rate of photogenerated electron-hole pairs inside the FeTiO₃ particles before reaching the surface [27]. However, undoped TiO₂ and C-TiO₂ show a weak activity (52.7%, 73.3% after 150 min). Compared to pure FeTiO₃ and C-TiO₂, C-TiO₂/FeTiO₃ composites were exhibited improved photocatalytic efficiency. The photocatalytic activity decreased gradually with increasing amount of FeTiO₃. This may be due to the fact that when the amount of FeTiO₃ is higher than 10 wt%, the surface of C-TiO₂ is covered by FeTiO₃, thus hindering the light absorption ability of C-TiO₂ and prevented from forming active heterojunctions [22]. The enhanced photocatalytic activity of C-TiO₂/FeTiO₃ composites was attributed to the high separation efficiency of photogenerated electron-hole pairs resulted from the formation of the C-TiO₂/FeTiO₃ heterojunction [22]. As shown in Fig.7a, MO self-photolysis without catalyst under sunlight irradiation was not observed after 150 min, indicating that MO is stable under sunlight irradiation. Figure 7b shows the MO degradation absorption spectra in the presence of CTFTO-10 under sunlight irradiation. The absorption intensity of MO significantly decreases with the increase in irradiation time and nearly disappears after 150 min. The kinetics of photocatalytic degradation of organic pollutants usually follows the Longmuir-Hinshelwood model. If the reactant concentration is low, the fallowing rate equation followed,

ln⁡lnCtC0=-kt,

where C₀ and C_t are the concentrations of the MO in solution at time 0 and t, respectively, k is apparent first-order rate constant, and t is the reaction time [27]. The plot of ln (C_t/C₀) versus irradiation time (t) is shown in Fig.7c, and the apparent rate constants were 0.0300, 0.0158, 0.0115, 0.0088, 0.0049, and 0.00007 for CTFTO-10, CTFTO-20, CTFTO-40, C-TiO₂, and FeTiO₃, respectively. This result shows that the introduction of FeTiO₃ could efficiently enhance the photocatalytic capability of C-TiO₂ under sunlight irradiation.

Fig.7

	Figure Option ViewDownloadNew Window
	Fig.7 MO photolysis and photocatalytic degradation over undoped TiO2, FeTiO3, C-TiO2, and C-TiO2/FeTiO3 composites a; UV-Vis absorption spectra for the degradation of MO in the presence of CTFTO-10 composite b and photocatalytic degradation kinetics of undoped TiO2, FeTiO3, C-TiO2, and CTFTO-10 composite for MO degradation under sunlight irradiation c

In order to evaluate the stability of CTFTO-10 during photocatalytic reaction, cycling experiments were also carried out by repeated decomposition of MO for three times under sunlight irradiation. As shown in Fig.8a, the photocatalytic activity of CTFTO-10 is high even with three cycles with a slight decrease, which could be due to the loss of the sample during the recycling experiments. Figure 8b shows the XRD patterns of the CTFTO-10 composite before and after photocatalytic reaction, and there is no observable structural difference in the samples before and after the reaction, indicating that the phase and structure of CTFTO-10 composite are stable during the photocatalytic reactions. These results indicate that the as-prepared composite catalysts are stable and reusable for photocatalytic reactions.

Fig.8

	Figure Option ViewDownloadNew Window
	Fig.8 Reusability study of CTFTO-10 composite for MO degradation under sunlight irradiation a; XRD patterns of the CTFTO-10 composite before and after photocatalytic reactions b

Figure 9a shows the photocatalytic degradation results in the presence of different scavengers under sunlight irradiation. The degradation efficiency of MO decreased slightly upon addition of AO compared with the absence of scavenger, which indicated the photogenerated holes are not the main active species for degradation of MO, nor the sources of^· OH. When BQ was used as a^· O₂^- scavenger, a dramatic change in the photocatalytic activity was observed compared with the absence of scavenger, confirming that the dissolved oxygen has a clear effect on the photodegradation process under sunlight irradiation. However, a similar change in the photocatalytic activity was observed upon the addition of IPA as a^· OH scavenger, which indicates that the^· O₂^- and^· OH radicals are the main reactive species in the C-TiO₂/FeTiO₃ system. Moreover, PL spectra of TA were used to disclose the formation of^· OH on the surface of photocatalyst, as shown in Fig.9b. The^· OH reacts with TA to form 2-hydroxyterephthalic acid, which emits a characteristic fluorescence signal at 425 nm and gradually increased with the irradiation time [33], which indicates the formation of^· OH under sunlight irradiation.

Fig.9

	Figure Option ViewDownloadNew Window
	Fig.9 Effect of different scavengers on degradation of MO in the presence of CTFTO-10 composite under sunlight irradiation a; · OH trapping PL spectra of CTFTO-10 composite in TA solution under sunlight irradiation b

On the basis of above results, a possible photocatalytic mechanism for the photodegradation of MO over C-TiO₂/FeTiO₃ photocatalyst under sunlight irradiation is proposed, as shown in Fig.10. The conduction band (CB) and valance band (VB) edge potentials of a semiconductor can be determined by using the following equation

ECB0=χ -E^e 12Eg

where χ is the absolute electronegativity of the semiconductor (for TiO₂, χ is 5.84 eV) [35], E^e is the energy of free electrons on the hydrogen scale (~4.5 eV), and E_g is the band gap energy of the semiconductor. According to the previous reports, the band gap of FeTiO₃ was reported to be in the range of 2.5-2.9 eV, and CB and VB redox potentials are ~0.0 and 2.6 eV, respectively [24, 28]. The calculated CB and VB redox potentials of the undoped TiO₂ are -0.245 eV and 2.925 eV. It is well known that nonmetal doping does not change the CB position of TiO₂. The VB position of C-TiO₂ can be calculated based on the calculated CB position of undoped TiO₂ and E_g of C-TiO₂ [35, 38], which was found to be 2.579 eV. Under sunlight irradiation, both C-TiO₂ and FeTiO₃ are excited followed by the formation of electrons and holes in the CB and VB, respectively. Then, the photogenerated electrons can easily migrate from the CB of C-TiO₂ to the CB of FeTiO₃ because the CB potential of C-TiO₂ (-0.245 eV vs. NHE) is more negative than that of FeTiO₃ (~0.0 eV vs. NHE) [27]. On the other hand, the VB potential of FeTiO₃ (2.6 eV vs. NHE) is more positive than that of C-TiO₂ (2.579 eV vs. NHE), so the photogenerated holes in the VB of FeTiO₃ can migrate to the VB of C-TiO₂. This is due to the synergistic effects at interface of the C-TiO₂ and FeTiO₃. Similar electron-hole transfer mechanism was reported for C-TiO₂/V₂O₅ composite [17]. This process could effectively improve the separation of photogenerated electron-hole pairs and decrease the possibility of charge recombination further improving the photodegradation efficiency. The electrons in the CB of FeTiO₃ reach the surface of photocatalyst to react with adsorbed oxygen to generate various reactive species (^· O₂^- and^· OH), while the holes on VB of C-TiO₂ react with adsorbed water to produce^· OH. These radicals are highly responsible for the degradation of organic pollutants [19, 44].

Fig.10

	Figure Option ViewDownloadNew Window
	Fig.10 A schematic mechanism of MO degradation over CTFTO-10 composite under sunlight irradiation

To further increase the photocatalytic activity, a double-heterojunctioned structure of CQD/C-TiO₂/FeTiO₃ was prepared by loading CQD onto the surface of the CTFTO-10 composite [45]. Figure 11 shows the morphologies of CQD, CTFTO-10, and CQD/CTFTO-10. Figure 11a shows the TEM image of the CQD. The as-prepared CQD was monodispersed with an average size of 10 nm. The SEM image of CTFTO-10 (Fig.11b) remains unchanged even after loading of CQD (CQD/CTFTO-10) (Fig.11c). Figure 12 shows the UV-Vis DRS, XRD, and PL spectra of the CTFTO-10 and CQD/CTFTO-10, respectively. CQD/CTFTO-10 displays strong absorption in the visible region compared with CTFTO-10 (Fig.12a), which implies that the photocatalyst may have higher photocatalytic activity for MO degradation and CQD may be playing an important role in utilizing sunlight [46]. The XRD pattern of CQD/CTFTO-10 shows similar diffraction peaks as CTFTO-10 (Fig.12b). As shown in Fig.12b, no diffraction peaks corresponding to CQD can be found, which is due to the small quantity of CQD in the composite and their poor crystallinity [47]. The PL intensity of CQD/CTFTO-10 was lower than that of CTFTO-10 (Fig.12c), which indicates that after the formation of a double heterojunction, the recombination rate of the photogenerated electron-hole pairs on the photocatalyst surface was suppressed and improved the photocatalytic activity.

Fig.11

	Figure Option ViewDownloadNew Window
	Fig.11 TEM image of CQD a, SEM images of CTFTO-10 composite b, CQD/C-TiO2/FeTiO3 composite c

Fig.12

	Figure Option ViewDownloadNew Window
	Fig.12 Diffuse reflectance UV-Vis spectra a, XRD pattern b, PL spectra c of CTFTO-10 composite and CQD/C-TiO2/FeTiO3 composite

The photocatalytic activity of CQD/CTFTO-10 composite was evaluated for the degradation of MO under sunlight irradiation, which is shown in Fig.13. CQD/CTFTO-10 composite exhibited the highest photocatalytic activity than that of CTFTO-10. MO was almost degraded in 120 min with CQD/CTFTO-10, whereas only 92% MO was degraded for CTFTO-10 within the same time. During the photocatalytic process, CQD can act as both electron acceptors and donors; therefore, the photogenerated electrons in CQD can be easily transferred to the surface of C-TiO₂/FeTiO₃, and the electrons on C-TiO₂/FeTiO₃ can also be transferred to CQD [46]. This transference efficiently suppresses the charge recombination and thus results in an enhanced photocatalytic activity. It is consistent with the PL results. Then, the photogenerated electron-hole pairs react with the adsorbed oxygen and water molecules to produce^· O₂^- and^· OH radicals, which subsequently cause the degradation of organic pollutants [19, 44].

Fig.13

	Figure Option ViewDownloadNew Window
	Fig.13 Photocatalytic degradation of MO over CTFTO-10 composite and CQD/CTFTO-10 composite