Influences of Sn/Cu single doping and CO-doping on the structure and photocatalytic property of anatase/rutile mixed crystal TiO2 nanomaterials under UV-visible light

Anatase/rutile mixed crystal TiO2 nanomaterials were prepared by sol-gel method and modified by Sn/Cu single doping and co-doping. Sn doping promotes the transformation from anatase to rutile, while Cu doping inhibits the phase transformation. The inhibition effect of Cu doping on phase transition is stronger than that of Sn doping. Sn or Cu doping reduces the recombination rate, and co-doping produces a synergistic effect on the inhibition of recombination. The photocatalytic experiment results show that the photocatalytic activity of Sn-TiO2 is higher than that of pure TiO2 owing to higher quantum efficiency and light source absorption. The first order reaction rate constant increases from 0.00904 min -1 for pure TiO2 to 0.01476 min -1 for Sn-TiO2. Unexpected, the photocatalytic activities of Cu-TiO2 and Sn/Cu-TiO2 are lower than that of pure TiO2. Although Cu doping improves the quantum efficiency, it reduces the absorption of ultraviolet region significantly, which is the key reason for the decline of their photocatalytic performance.


Introduction
Degradation of pollutants by TiO 2 photocatalyst materials has a broad application prospect in the field of environmental protection [1][2][3][4][5]. The visible light utilization and quantum efficiency of TiO 2 can be improved by modification [6][7][8][9][10]. Ion doping is simple and effective way to modify the photocatalytic activity, which has been widely concerned [11][12][13]. Udayabhanu et al. [13] found that Cu doping improved the utilization of visible light and the separation of photogenerated electron-hole pair, promoting the photocatalytic activity. The co-doping modification may produce a synergistic effect and achieve higher photocatalytic activity than that of single element modification [14][15][16]. Georgieva et al. [16] prepared B and N co-doped TiO 2 nanotubes. The synergistic effect of B and N co-doping resulted in the maximum quantum efficiency, thus codoped TiO 2 showed best photocatalytic activity.
In the present work, anatase/rutile mixed crystal TiO 2 was prepared, and the effects of Sn/Cu single doping and co-doping on the structure and photocatalytic performance of mixed crystal TiO 2 were investigated in detail.

Sample preparation
Solution A was obtained by adding anhydrous ethanol and tetrabutyl titanate in beaker with a volume ratio of 3:2. Deionized water, glacial acetic acid and anhydrous ethanol were mixed with a volume ratio of 4:4:15 to gain solution B, which was dropped into solution A to form a gel. After drying, it was heat treated at 550 ℃ for 1 h to obtain pure TiO 2 , which is labeled as PT. Adding a certain amount of Cu(NO 3 ) 2 ·3H 2 O or SnCl 4 ·5H 2 O into solution B, keeping the molar ratios of Ti/Sn and Ti/Cu are 1:0.05 and the rest steps are unchanged to prepare 5%Sn-doped TiO 2 and 5%Cu-doped TiO 2 , which are labeled as ST and CT, respectively. Meanwhile, Sn/Cu codoped TiO 2 with a molar ratio of Ti/Sn/Cu=1:0.025:0.025 marked as SCT can be obtained by adding moderate Cu(NO 3 ) 2 ·3H 2 O and SnCl 4 ·5H 2 O simultaneously.

Sample characterization
A DX-2700 X-ray diffractometer was used to analyze the crystal structure (XRD). A FEI-Inspect F50 scanning electron microscope and a FEI-Tecnai G2 F20 transmission electron microscope were used to observe the morphology (SEM and TEM). An XSAM800 multifunctional surface analysis system was used to analyze the element composition and valence state (XPS). A F-4600 fluorescence spectrometer was used to analyze the photoinduced electronhole recombination rate (PL). A UV-3600 ultraviolet-visible photometer was used to test the optical absorption (DRS).

Photocatalytic experiment
The photocatalytic performance of samples was investigated by the degradation rate of Rhodamine B solution (RhB). 200 mL RhB solution with a concentration of 10 mg/L and 0.1 g sample powder were mixed. After stirred for 30 min in dark, the mixture was irradiated by a 250 W xenon lamp. The samples were taken every 20 min to measure its absorbance at 553 nm. RhB degradation rate was calculated by the formula (A 0 -A t )/A 0 ×100%, where A 0 and A t are the initial and t time absorbance.  211) and (220) crystal planes of rutile structure. PT is a mixed crystal structure composed of anatase and rutile, and the content of anatase is 69.3% and rutile is 30.7%. The rutile diffraction peaks intensity of ST is higher than that of PT, and the rutile content of ST is 85.6%, indicating that Sn doping facilitates the transformation from anatase to rutile phase. Since Sn 4+ radius (0.069 nm) is slightly larger than Ti 4+ radius (0.0605 nm), Sn 4+ ions enter into TiO 2 lattice and replace Ti 4+ ions, forming lattice defects and introducing oxygen vacancies, making Ti-O bond more prone to fracture and promoting the phase transformation [21,22]. The CT pattern is full of anatase diffraction peaks, and there is no rutile related diffraction peak, which suggests that Cu doping inhibits the phase transformation from anatase to rutile. Cu element may be dispersed on the surface of TiO 2 in the form of oxide, which obstructs the migration of Ti and O atoms at the interface thus retards the nucleation and growth of rutile [23]. SCT is a mixed crystal structure, and the content of rutile is 7.0%, which is lower than that of PT (30.7%), indicating that the inhibition of Cu doping is stronger than the promotion of Sn doping, thus Sn/Cu co-doping inhibits the phase transformation. 20 30 The average grain size of samples were calculated using Scherrer formula [24] and the results are shown in Fig. 2. The grain sizes of doped TiO 2 are less than pure TiO 2 , implying that both single doping and co-doping can refine the grains.  Fig. 3 presents the SEM images of samples. In Fig. 3a, it is observed that the particle morphology of PT is granular with a certain degree of agglomeration, and the agglomerate sizes are ranged from tens to hundreds of nanometers. There is no obvious difference in the morphology of doped samples, and all of them are composed of granular particles and agglomerations. Fig. 4 shows the TEM and HRTEM images of PT (a, c) and SCT (b, d). The single particle size of PT is about 20-30 nm in Fig. 4a. In Fig. 4c, the crystal plane spacing 0.346 nm can be ascribed to the anatase (101) crystal plane and the crystal plane spacing 0.249 nm corresponds to the rutile (101) crystal plane [8], which indicates that PT forms anatase and rutile mixed crystal structure. It is observed in Fig. 4b that the single particle size is about 20-30 nm, and the interplanar spacing 0.358 nm corresponds to the anatase (101) crystal plane.   Fig. 5 shows the XPS spectrum of SCT. The characteristic peaks of C 1s, Ti 2p, Sn 3d, Cu 2p and O 1s appear is the full spectrum (Fig. 5a), indicating the presence of C, Ti, Sn, Cu and O elements in the sample. The high-resolution spectrum of Ti 2p is shown in Fig. 5b. Ti 2p consists of four peaks located at 463.4 eV, 457.9 eV and 464.7 eV, 459.0 eV corresponds to Ti 3+ 2p 1/2 , Ti 3+ 2p 3/2 and Ti 4+ 2p 1/2 , Ti 4+ 2p 3/2 , indicating that Ti element exists +3 and +4 valences [25,26]. Fig. 4c shows the high-resolution spectrum of Sn 3d. The characteristic peaks of Sn 4+ 3d 5/2 and Sn 4+ 3d 3/2 appear at 486.0 eV and 494.7 eV, respectively, indicating that Sn exists as Sn 4+ [27]. It can be seen from Fig. 5d that Cu 2p consists of four peaks located at 933.0 eV, 942.3eV, 952.9 eV and 962.3eV, suggesting that Cu element exists as Cu 2+ and Cu + [12,28,29]. The characteristic peaks in Fig. 5e are located at 529.5 eV and 530.9 eV, corresponding to the lattice oxygen (O 2-) and surface hydroxyl (OH -) [30], respectively.

Optical property
The photons will be released when photogenerated electrons and holes recombine, which is related to the PL peak intensity. Therefore, the lower PL peak intensity, the lower the recombination rate of photogenerated electrons and holes [31]. Fig. 6 presents the PL spectra of samples. The PL peak intensity of ST is lower than that of PT, indicating that Sn doping inhibits the recombination. The replacement of Ti 4+ by Sn 4+ introduces lattice defects and O vacancies, which capture the photogenerated charges and reduce the recombination [32]. Remarkably, the PL peak intensity decreases significantly after Cu doping. Cu oxides may be dispersed on the surface of TiO 2 , forming TiO 2 /CuO and TiO 2 /Cu 2 O semiconductor composites. Since the conduction band of TiO 2 is higher than that of Cu oxides, the photogenerated electrons will migrate from TiO 2 conduction band to Cu oxides conduction band after being excited by light, thus reducing the recombination rate of photogenerated electrons and holes in TiO 2 [33]. SCT shows the lowest PL peak intensity, indicating that Sn and Cu co-doping produces a synergistic effect on inhibiting recombination thus exhibits the highest quantum efficiency. The UV-vis absorption spectra of samples are depicted in Fig. 7. The absorption of ST in UV region is higher than that of PT. Meanwhile, the absorption edge of ST shows a red shift compared with PT, indicating that Sn doping enhances the utilization of light source. The most noteworthy thing is that the absorption in UV region of CT and SCT decreases sharply compared to PT, which may be due to the fact that Cu oxides disperses on TiO 2 surface, preventing the absorption of light source [34]. Marked reduction of light source absorption by Cu doping is detrimental to the photocatalytic performance.

Photocatalytic activity
The photocatalytic performance of samples was evaluated by the degradation degree of RhB and the results are shown in Fig. 8. The degradation degree of ST is 78.5% after 100 min, which is higher than that of PT (58.9%), indicating that Sn doping improves the photocatalytic performance of TiO 2 . The degradation degree of CT is 8.4%, which is significantly lower than that of PT, indicating that Cu doping suppresses the photocatalytic performance. The degradation degree of SCT is 11.1%, showing that the photocatalytic activity is still reduced after co-doping. The first-order reaction rate constants of PT, ST, CT and SCT are 0.00904 min -1 , 0.01476 min -1 , 0.00089 min -1 and 0.00120 min -1 , respectively. ST exhibits the fastest reaction rate and the reaction rate of CT and SCT are less than PT, which is consistent with the results of degradation degree.
It can be concluded from optical measurement results that Sn doping is advantageous to enhancing the light utilization and quantum efficiency, thereby improving the photocatalytic performance. On the other hand, although Cu doping is beneficial to reducing the recombination of photogenerated electrons and holes, it does not elevate the photocatalytic activity. The absorption of ultraviolet light is drastically reduced after Cu doping, which may be caused by Cu oxides dispersing on the surface of TiO 2 , preventing the light absorption. This is the critical reason for the decline of photocatalytic performance after Cu doping. The UV absorption of SCT is slightly higher than that of CT, as a result, the photocatalytic activity of co-doping TiO 2 is higher than that of Cu-doped TiO 2 .

Conclusions
Pure and Sn/Cu doped TiO 2 were prepared by sol-gel route and characterized by XRD, SEM, TEM, XPS, PL and DRS. The results show that Sn doping promotes the phase transformation, while Cu doping inhibits the phase transformation. The inhibitory effect of Cu doping is stronger than the promotion of Sn doping, and co-doping still shows the inhibitory effect. Both Sn/Cu single doping and co-doping reduce the recombination rate of photogenerated electron-hole pairs. The absorption of light source is increased by Sn doping, however, the absorption of CT and SCT in UV region drop sharply. The photocatalytic experiment results show that Sn doping improves the photocatalytic performance of TiO 2 . On the other hand, Cu doping and Sn/Cu co-doping suppress the photocatalytic performance of TiO 2 due to their poor absorption in UV region.