Cerium Synchronous Doping in Anatase for Enhanced Photocatalytic Hydrogen Production from Ethanol-Water Mixtures

Cerium element with a unique electric structure can be used to modify semiconductor photocatalysts to enhance their photocatalytic performances. In this work, Ce-doped TiO2 (Ce/TiO2) was successfully achieved using the sol-gel method. The structural characterization methods confirm that Ce was doped in the lattice of anatase TiO2, which led to a smaller grain size. The performance test results show that the Ce doped in anatase TiO2 significantly enhances the charge transport efficiency and broadens the light absorption range, resulting in higher photocatalytic performance. The Ce/TiO2 exhibited a photocurrent density of 10.9 μA/cm2 at 1.0 V vs. Ag/AgCl, 2.5 times higher than that of pure TiO2 (4.3 μA/cm2) under AM 1.5 G light. The hydrogen (H2) production rate of the Ce/TiO2 was approximately 0.33 μmol/h/g, which is more than twice as much as that of the pure anatase TiO2 (0.12 μmol/h/g). This work demonstrates the effect of Ce doping in the lattice of TiO2 for enhanced photocatalytic hydrogen production.


Introduction
H 2 energy, with its high calorific value and environmental friendliness, has been regarded as an alternative to unsustainable fossil fuels with the most potential [1][2][3]. Solar to H 2 conversion is a promising technique for solving the energy crisis and environmental problems caused by continuously consuming nonrenewable fossil fuels [4,5]. Since Fujishima and Honda first discovered the photoelectrochemical splitting of water into H 2 and O 2 using TiO 2 photoanode [6][7][8][9], TiO 2 has been widely applied in photocatalysis due to the fact of its low cost, stability, and nontoxicity [10][11][12]. However, TiO 2 only reacts to ultraviolet light due to the fact of its wide band gap of 3.2 eV, resulting in low quantum efficiency. In addition, the high-rate recombination of photogenerated electron/hole pairs causes the practical application of TiO 2 for photocatalytic H 2 production to be strictly confined. To address these drawbacks and boost the solar energy utilization efficiency of TiO 2 , considerable efforts have been devoted. Many strategies, for instance, metal doping, such as Ta and Fe doping [13,14]; nonmetal doping, such as N and C doping [15,16]; surface dye sensitization [17]; and noble Ag and Au doping [10,18], is often employed to modulate the electronic structure and photocatalytic activity of TiO 2 . In particular, introducing rare earth elements into TiO 2 can effectively improve the electron-hole separation and extend the visible light response of TiO 2 [19,20]. Among rare earth elements, cerium (Ce) shows the variable valence states Ce 3+ /Ce 4+ with different electronic structures (4f 1 5d 0 and 4f 0 5d 0 , respectively), which result in the facile formation of oxygen vacancies. Thus, the shuttling of electrons between two valence states largely facilitates the charge separation and the following photocatalytic reactions [21,22]. However, most research on Ce-doped titanium dioxide focuses on photocatalytic degradation, with few studies on photocatalytic water splitting.
In this work, a simple sol-gel method achieved Ce synchronous doping in the lattice of TiO 2 . The morphology and structure of the pure TiO 2 , Ce-doped TiO 2 (Ce/TiO 2 ), and CeO 2 -mixed TiO 2 (CeO 2 -TiO 2 ) samples were observed by X-ray diffraction (XRD), scanning electron microscopy (SEM), and transmission electron microscopy (TEM). The PEC performances of TiO 2 , Ce/TiO 2 , and CeO 2 -TiO 2 were comprehensively investigated. The electron capture center formed by Ce doping into the TiO 2 lattice dramatically improves the separation efficiency of photogenerated electrons and holes. In addition, the narrowed band gap of the Ce-doped TiO 2 shows excellent visible light absorption and photocurrent response. The Ce/TiO 2 samples realized high photocurrent density and incident photonto-current efficiency (IPCE) values thanks to the Ce doping. The Ce doping also hugely improved the photocatalytic H 2 production of the TiO 2 electrode. This work offers a practical strategy and significant reference for preparing and understanding efficient visiblelight-activated rare-earth-doped photocatalysts.

Structure and Morphology
The surface morphology of the pure TiO 2 , Ce-doped TiO 2 (Ce/TiO 2 , 0.5% Ce), and CeO 2 -mixed TiO 2 (CeO 2 -TiO 2 ) samples were observed by SEM, and their crystal structures were identified by XRD measurement, as shown in Figure 1. The elemental mapping results of the as-prepared Ce/TiO 2 sample demonstrated a uniform distribution of Ce in TiO 2 , as shown in Figure 1d. As shown in the SEM and EDS mapping images, Ce/TiO 2 consists of TiO 2 nanoparticles with a homogeneous distribution of Ce elements. The relative amount of the Ce element in Ce/TiO 2 is close to that in CeO 2 -TiO 2 , as shown in Table S1. The crystal structures of these obtained samples were identified by XRD, as shown in Figure 1e,f. The pure TiO 2 , CeO 2 -TiO 2 , and Ce/TiO 2 powders comprised the anatase phase dominantly. The samples of pure TiO 2 and CeO 2 -TiO 2 still had a rutile phase with a peak at 27.4 • corresponding to the (110) plane of the rutile TiO 2 . However, the growth of the rutile phase in the pure TiO 2 was inhibited after doping with Ce. In addition, the peak intensity of the anatase TiO 2 in the Ce-doped TiO 2 samples decreased when the Ce content increased. Excessive Ce ions stay at the particle surface or the grain boundaries, suppressing the anatase crystal growth [23,24]. From the high-resolution plots of the prominent peak (101) at 25.48 • , as shown in Figure 1b, the crystal sizes of the pure TiO 2 , 0.5%Ce/TiO 2 , 3%Ce/TiO 2 , and 5%Ce/TiO 2 were calculated to be 13.92, 12.03, 12.09, and 12.14 nm, respectively, according to the Scherrer formula [22]. The peaks of the anatase (101) crystal plane in the Ce-doped TiO 2 samples shifted towards a lower diffraction angle compared to those of the pure TiO 2 and CeO 2 -TiO 2 . The interplanar spacing of the (101) crystal plane increased due to the Ce doping [24]. No prominent characteristic peaks of CeO 2 were observed in the Ce-doped TiO 2 samples due to the low doping contents.
As shown in Figure 2, the internal microstructures of the pure TiO 2 , Ce/TiO 2 (0.5%Ce/ TiO 2 ), and CeO 2 -TiO 2 were explored by TEM, HRTEM, SAED, and EDS mapping, respectively. As shown in the TEM images, the Ce/TiO 2 consisted of TiO 2 nanoparticles with a homogeneous distribution of Ce elements. The pure TiO 2 mainly consisted of anatase nanoparticles, as shown in Figure 2a Figure 2g shows both the anatase plane and the CeO 2 plane. From the EDS mapping image, as shown in Figure 2h, it is clear that the Ce elements are distributed unevenly in the composite structure of CeO 2 -TiO 2 . However, in the sample of the Ce-doped TiO 2 (0.5%Ce/TiO 2 ), the nanoparticle size was smaller than that of the composite CeO 2 -TiO 2 , which is consistent with the results of the XRD and SEM. The HRTEM Image (Figure 2j) of Ce/TiO 2 shows that the lattice spacing of the anatase (101) plane increased to 0.366 nm. In addition, several bright spots distributed in the anatase lattice are highlighted by the red circles, indicating Ce atoms doping into the lattice of the TiO 2 . The SAED patterns in Figure 2k also confirm the single crystallinity of the anatase TiO 2 in the Ce-doped TiO 2 . The elemental mapping results shown in Figure 2l further verify the uniform distribution of Ce element in the lattice structure of TiO 2 in Ce/TiO 2 . Therefore, the sol-gel method successfully doped the Ce in the lattice of TiO 2 .

Chemical Composition and Surface Information
For obtaining the chemical composition and valence state of the Ce-doped TiO 2 , X-ray photoelectron spectroscopy (XPS) measurements were conducted, as shown in Figure 3. The pure TiO 2 , Ce/TiO 2 , and CeO 2 -TiO 2 were all composed of O, Ti, and C, as shown in the survey spectra ( Figure 3a). The Ti 2p peaks at 458.28 eV and 463.98 eV are attributed to Ti 2p 3/2 and Ti 2p 1/2 , respectively, indicating the Ti 4+ state in the Ce/TiO 2 , as shown in Figure 3b. A slight shift in the Ti 2p in the Ce-doped TiO 2 towards a lower value compared to that of pure TiO 2 could be due to the interaction between Ti 4+ and Ce 3+ or Ce 4+ species [25]. In Figure 3c, the O 1s peak at 529.73 eV corresponds to the lattice oxygen of TiO 2 , while the peak at 531.48 eV corresponds to the surface-adsorbed oxygen species [26]. The O 1s spectrum of the Ce/TiO 2 had a smaller peak area at 532.78 eV compared to that of pure TiO 2 , which indicates a reduction in oxygen vacancies. In addition, the O 1s peak slightly shifted towards the lower binding energy region due to the Ce doping. The Ce 3d spectra (Figure 3d) show eight fitting peaks, where a and b correspond to the two spin orbitals (Ce 3d 5/2 and Ce 3d 3/2 ) in the Ce/TiO 2 , respectively [27]. The peaks a 2 and b 2 are attributed to Ce 3+ , while the others correspond to Ce 4+ . The relative content of Ce ions can be calculated from the ratio of the fitting peak area. The relative content of Ce 3+ in the Ce/TiO 2 (30.4%) was higher than that of the CeO 2 -TiO 2 (24.6%), indicating that more Ce 3+ ions were retained in the synchronous doping process. The higher content of Ce 3+ in Ce/TiO 2 plays a key role in the formation of oxygen defects and improves the redox conversion between Ce 3+ and Ce 4+ [28]. The UV-Vis absorbance spectra in Figure 4a show the redshift of the absorption edge to the visible region in TiO 2 after Ce doping, which may be due to the Ce 4f intermediate energy level inside the band gap of TiO 2 [25]. The Tauc plot extracted from the UV-Vis absorbance spectra exhibits the band gap of the samples, as shown in Figure 4b. The band gaps of the pure TiO 2 , CeO 2 -TiO 2 , and Ce/TiO 2 were 2.96 eV, 2.91 eV, and 2.15 eV, respectively.
For further study of the separation of the photogenerated electron/hole pairs, fluorescence (FL) spectroscopy was conducted, as shown in Figure S1. The luminous signal was generated in the region of 375-450 nm, with the longest signal at 420 nm, indicating the generation of * OH radicals in the reaction process. Generally, a higher intensity of FL suggests a higher recombination of the photogenerated carriers [29]. Pure TiO 2 had a stronger spectral signal than that of the Ce-doped TiO 2 , indicating the high recombination rate of the photogenerated carriers in the pure TiO 2 . The Ce doping effectively improved the charge separation in the TiO 2 . However, the FL intensities of 3% Ce/TiO 2 and 5% Ce/TiO 2 increased compared to that of the 0.5% Ce/TiO 2 . The excessive Ce doping could form a new electron-hole recombination center, resulting in a lowered efficiency of charge separation. Therefore, 0.5% of Ce doping in TiO 2 could provide better efficiency for charge separation and photocatalytic reactions compared to pure TiO 2 and other Ce-doped TiO 2 samples with high Ce doping contents.

Photoelectrochemical Performance
The PEC performance of these as-prepared samples was further investigated. Under chopping irradiation, a linear sweep voltammetry (LSV) measurement was carried out under AM 1.5 G light, as presented in Figure 5a. Ce/TiO 2 shows a photocurrent density of 10.9 µA/cm 2 at 1.0 V vs. Ag/AgCl, which is 2.5 and 2.4 times higher than that of pure TiO 2 (4.3 µA/cm 2 ) and CeO 2 -TiO 2 (4.5 µA/cm 2 ), respectively. The highly increased photocurrent of Ce/TiO 2 mainly originates from the efficient charge separation. Ce doping in TiO 2 enhances the efficiency of photogeneration and separation of electron-hole pairs, resulting in a high photocurrent response of Ce/TiO 2 . Figure S2 presents the chopped LSV curves of the as-prepared TiO 2 samples, which were modified with different Ce loadings or were annealed at different temperatures. Chronoamperometry measurement was used to investigate the chemical stability of the samples at 0.8 V vs. Ag/AgCl, as presented in Figure 5b. The CeO 2 -TiO 2 showed a higher photocurrent density than the pure TiO 2 , which resulted from the heterojunction between TiO 2 and CeO 2 . The maximum photocurrent densities of the pure TiO 2 , CeO 2 -TiO 2 , and Ce/TiO 2 were 3.2, 4.5, and 8.0 µA/cm 2 at 0.8 V vs. Ag/AgCl, respectively. These results indicate that Ce doping into the crystal lattice of TiO 2 is beneficial to the separation of photogenerated electrons and holes. Ce ions in TiO 2 form an intermediate energy level, which results in the formation of oxygen vacancies. Under light radiation, the absorbed oxygen on the surface of the photocatalyst generates hydroxyl radicals and superoxide radicals, improving the photocatalyst's activity. The Ce/TiO 2 sample exhibited steady photocurrent densities during the long cycling, implying that its photostability significantly improved.
The onset potential of the photocurrents indicates the photocatalytic activity and the flat band potential of the photoanodes. A high onset potential implies a low utilization efficiency of solar energy [30]. As shown in Figure 6, the onset potential of the Ce/TiO 2 (0.0 V) was more negative compared to that of the pure TiO 2 (0.159 V) and CeO 2 -TiO 2 (0.103 V). The low onset potential of the Ce/TiO 2 indicates an efficient carrier transfer process in the Ce-doped TiO 2 [31]. The photocurrent densities versus the monochromatic light of the samples were measured in electrochemical noise (ECN) mode, which is a nondestructive and in situ monitoring technique to investigate the spontaneous electrochemical reactions of photoanodes [16,32], as shown in Figure 7a. The photocurrent density of the Ce/TiO 2 was slightly lower than that of the pure TiO 2 under the UV light region. It is worth noting that the Ce/TiO 2 showed visible light absorption up to 500 nm, while the pure TiO 2 and CeO 2 -TiO 2 exhibited no obvious response under the visible light region. The successful Ce doping highly improves the visible light utilization efficiency of TiO 2 . The incident photonto-current efficiencies (IPCEs) spectra were obtained from the photocurrent-wavelength curves, as shown in Figure 7b. The Ce/TiO 2 had the highest IPCE value of 0.10% in the region of 350 nm and an obvious visible light response. The Ce/TiO 2 still exhibited an IPCE value of 0.02% at 400 nm. The photocurrents and IPCE versus monochromatic light of Ce/TiO 2 under applied voltages are shown in Figure 7c. The photocurrent response of the Ce/TiO 2 increased with bias, which is consistent with the results of the LSV curves. The charge transport in the pure TiO 2 , CeO 2 -TiO 2 , and Ce/TiO 2 was also studied by electrochemical impedance spectroscopy (EIS), as shown in Figure 8c. The Ce/TiO 2 sample shows a smaller arc in the Nyquist plot, indicating a lower charge transfer resistance at the electrode interface [36]. Ce doping introduces a new electronic state within the band gap of TiO 2 near the conduction band, facilitating the charge separation and reducing the recombination rate of photogenerated electron/hole pairs. Furthermore, the Mott-Schottky (M-S) measurement was used to determine the flat band potential (V fb ) of these samples, as illustrated in Figure 8b. The curve slopes of the three samples are all positive, indicating their n-type semiconductor properties. The flat band potential can be regarded as the conduction band (CB) potential of n-type semiconductors, which is calculated by intercepting the slopes with the potential axis according to the M-S equation [25,34]. The values of V fb of pure TiO 2 , CeO 2 -TiO 2 , and Ce/TiO 2 are about 0.0, −0.13, and −0.16 V vs. RHE, respectively. The negative shift in the flat band potential of the Ce/TiO 2 compared to the pure TiO 2 implies a decrease in the transfer energy barrier of the interfacial electrons and the charge transfer resistance after Ce doping in TiO 2 [35]. In addition, the slope of the M-S plot of the Ce/TiO 2 is significantly smaller than that of the pure TiO 2 , indicating an increase in the charge carrier density according to the M-S equation.
The band gap energy Eg of these samples can be calculated from the IPCE spectra by a Tauc plot of (IPCE % × hv) 1/2 versus photon energy (hv) [16,[33][34][35], as illustrated in Figure 8a. The band gaps of the pure TiO 2 , CeO 2 -TiO 2 , and Ce/TiO 2 were 3.23, 3.24, and 2.73 eV, respectively. The band gap of the pure TiO 2 is consistent with the reported data. The Ce/TiO 2 had a lower Eg than pure TiO 2 owing to the formation of oxygen defect levels above the valence band of TiO 2 . Thus, Ce doping in TiO 2 narrows the band gap and enhances the visible light absorption of TiO 2 .

PC H 2 Evolution Performance and Mechanism for Water Splitting under Solar Light
The photocatalytic H 2 evolution of the pure TiO 2 , CeO 2 -TiO 2 , and Ce/TiO 2 was evaluated under simulated solar light, as shown in Figure 9a. The H 2 evolution rate of Ce/TiO 2 was approximately 0.33 µmol/h/g, which is more than twice that of the pure TiO 2 (0.12 µmol/h/g) and CeO 2 -TiO 2 (0.16 µmol/h/g). Moreover, Ce/TiO 2 also showed superior photostability for the photocatalytic H 2 evolution over 10 h. The good performance in the water splitting of the Ce/TiO 2 resulted from the narrow band gap and the effective separation of the photogenerated carriers. Ce doping in the lattice of TiO 2 highly improves the photocatalytic performance of TiO 2 . The unique energy level structure of the Ce element provides doping levels in TiO 2 , facilitating electron transfer and charge separation.
Based on the above results, the mechanism of Ce doping in TiO 2 for an improved photocatalytic performance was proposed. The energy band diagram of the pure TiO 2 , CeO 2 -TiO 2 , and Ce/TiO 2 is illustrated in Figure 9b. Ce doping in the lattice of TiO 2 narrows the band gap of TiO 2 , resulting in a visible-light response. The negatively shifted conduction band in Ce/TiO 2 improves the carrier transfer process and reduces the recombination of photogenerated electron/hole pairs. Therefore, Ce/TiO 2 exhibits a high photocatalytic performance for H 2 evolution from water splitting.

Materials and Synthesis Methods
In the experiment, tetrabutyl titanate (TBOT), ethanol absolute, acetic acid, cerium nitrate hexahydrate (Ce(NO 3 ) 3 ·6H 2 O), and cerium oxide (CeO 2 ) were all of analytical grade and purchased from commercial sources. Deionized water was used in all experiments.
The process of sample preparation was conducted using the sol-gel method. Firstly, the mixed solution of tetrabutyl titanate (8.5 mL) and ethanol absolute (20 mL) was prepared under magnetic stirring and recorded as solution A. Solution B, which contained 8.5 mL of deionized water, 7.5 mL of acetic acid, 20 mL of ethanol absolute, and a specific amount of Ce(NO 3 ) 3 ·6H 2 O, was prepared. Then, solution B was slowly dropped into the vigorously stirred solution A at a rate of one drop every 2 s, resulting in the formation of TiO 2 sol, which was further heated at 70 • C for 4 h and dried at 70 • C for 10 h. Finally, the above gel was annealed at 500 • C for 5 h in the air. The obtained powders were marked as X% Ce/TiO 2 , where "X" represents the theoretical molar percentage of Ce in the product (0.5, 3.0, and 5.0%).
For convenience, the Ce/TiO 2 label mentioned below stands for the sample of 0.5% Ce/TiO 2 with the best performance by default. TiO 2 without Ce was prepared under the same conditions. For comparison, the TiO 2 powders obtained by the sol-gel method were mixed with commercial CeO 2 powders in a certain proportion (0.5% Ce), and the resulting powders were designated as CeO 2 -TiO 2 . The sample synthesis diagram is illustrated in Scheme 1.

Electrochemistry (EC) and Photoelectrochemical (PEC) Testing
All EC and PEC performances of the Ce/TiO 2 , CeO 2 -TiO 2 , and pure TiO 2 were carried out in a standard three-electrode configuration with a side quartz window in which a mixture solution of 0.1 M Na 2 SO 4 and ethanol (1:1 in volume, pH = 6.5) was the supporting electrolyte, and it was deaerated by bubbling high-purity Ar for 15 min before the EC/PEC measurements. A platinum plate and an Ag/AgCl electrode (saturated KCl) served as the counter electrode and reference electrode, respectively.
The working electrodes were fabricated on fluorine-doped conductive (FTO) glass plates (30 × 35 × 2.2 mm) for the PEC measurements. After being sonicated in soap suds and acetone, the FTO glass plates were rinsed with deionized water and then dried in an N 2 atmosphere. A mixed solution, including 5 mg of the as-prepared photocatalyst powders, 1 mL of absolute ethanol, and 20 µL of 5 wt.% naphthol solution, was sonicated for 90 min and used for spinning casting onto the conductive surface of the FTO glass plates with a speed of 3000 r/min. The exposed area of the FTO glass was controlled to 1.0 cm 2 (10 × 10 mm) by utilizing tape. After, the electrodes were dried at 80 • C for 2 h. The linear sweep voltammetry (LSV) and chronoamperometry (I-t) curves of the samples were carried out in a three-electrode side window electrolytic cell, which was measured on a CHI760E electrochemical workstation under a high uniformity integrated Xenon light source (PLS-FX300HU, Beijing Perfectlight, Beijing, China) with an AM 1.5 G filter (100 mW/cm 2 ).
An LSV measurement was typically carried out over a potential range of −0.2 to 1.0 V (vs. Ag/AgCl), and the scanning speed was 5 mV s −1 with the light cut off by a 5 s −1 shutter. An I-t measurement was performed under alternating illumination at 0.8 V (vs. Ag/AgCl). During the test, the working electrode was guaranteed to have an illuminated area of 1 cm 2 and completely submerged in the electrolyte. The calculated voltage can be changed to a reversible hydrogen electrode (RHE) scale using the Nernst equation: where E 0 Ag/AgCl = 0.1976 V vs. Ag/AgCl at room temperature. The incident photo-to-current efficiency (IPCE) test system was composed of an electrochemical workstation (CS350H, CorrTest, Wuhan, China), a 300 W xenon lamp light source (PLS-SXE300D, Beijing Perfectlight, Beijing, China), and a grating monochromator (7ISU, SOFN Instruments Co., Ltd., Beijing, China) with filters to remove higher order diffraction, which is calculated using the following equation: where J is the photocurrent density measured at a specific wavelength, λ is the specific wavelength, and I 0 is the intensity of the incident light. Electrochemical impedance spectroscopy (EIS) and Mott-Schottky (M-S) plots were measured in the dark, using an electrochemical workstation (Squidstat Plus, Admiral Instruments, Tempe, AZ, USA). The fitted circuit can be obtained from the EIS curve, where the CPE element is calculated. The impedance of the CPE in an AC circuit is: where σ is the prefactor of the CPE, ω is the angular frequency, m is the CPE index (0 ≤ m ≤ 1), and j is an imaginary number j = √ −1 ; if m = 1, then the CPE denotes the ideal capacitor C. The electrochemical analyzer works over a frequency range of 0.01 to 100,000 Hz, with voltage increments of 0.005 V and an AC amplitude of 10 mV. In a typical M-S measurement, the working electrodes were tested at 500, 1000, 1500, 2000, and 2500 Hz.

Photocatalytic (PC) Performances Testing
The PC splitting of the ethanol-water mixtures to generate the H 2 measurement was carried out in the MCP-WS1000 photochemical workstation (Beijing Perfectlight Technology Co., Ltd., Beijing, China) equipped with a 50 mL quartz-made reactor under artificial solar irradiation. Specifically, 20 mg of as-prepared photocatalysts was ultrasonically dispersed (30 min) into a solution consisting only of 15 mL ethanol and 15 mL water (pH = 7), which were deaerated by vacuuming for 10 min before the H 2 evolution measurements. The full spectrum source was simulated sunlight consisting of 9 LED lamps (365, 385, 420, 450, 485, 535, 595, and 630 nm and one white light LED, which was 420-750 nm), and the power of the total light irradiation was approximately 100 mW/cm 2 . The visible light source consisted of 9 white light LED lamps (420-750 nm) with a total visible light irradiation power of 100 mW/cm 2 . The temperature of the reaction was controlled by circulating condensed water at 5 • C throughout the reaction process using a water-cooling system. The produced hydrogen was calculated by the external standard method using the peak area obtained by the PLD-CGA1000 composite gas analyzer (Beijing Perfectlight Technology Co., Ltd., Beijing, China).

Materials Characterization
The morphology of the samples was observed by a scanning electron microscope with an operating voltage of 10 kV (SEM, Apreo S LoV ac, Thermo Fisher Scientific, Waltham, MA, USA).
The transmission electron microscopy and selected area electron diffraction images were obtained from an FEI Talos F200s transmission electron microscope, operated at 200 kV (TEM, SAED, Thermo Fisher Scientific, Waltham, MA, USA).
High-resolution transmission electron microscopy was carried out on an FEI Titan Themis Z G3 30-300 spherical aberration-corrected transmission electron microscope, operated at 300 kV, equipped with both image and probe aberration (HRTEM, Thermo Fisher Scientific, Waltham, MA, USA).
X-ray photoelectron spectroscopy was used to characterize the atomic composition and state at the surface of the samples, using Al Ka rays as the excitation source (XPS, Thermo Scientific K-Alpha, Waltham, MA, USA).
The photoluminescence spectra were acquired using a fluorescence spectroscopy test system (PL, FLS980, Edinburgh Instruments Ltd., Livingston, UK).

Conclusions
Ce synchronous doping in the lattice of anatase TiO 2 was successfully achieved using a facile sol-gel method. The Ce doping suppressed the rutile phase and crystal growth of the anatase TiO 2 , leading to smaller grain size and a higher specific surface area compared to pristine TiO 2 . Ce doping in the lattice structure of TiO 2 was also demonstrated by HRTEM and XPS. The charge recombination in the TiO 2 was successfully reduced by Ce doping. The Ce-doped TiO 2 showed a negative shift in the flat band potential and a lower carrier transport resistance. In addition, Ce doping narrowed the band gap of TiO 2 , leading to a visible light response. Therefore, Ce-doped TiO 2 exhibits an enhanced photocurrent density and an excellent photocatalytic performance. The Ce-doped TiO 2 showed superior photocatalytic performance for H 2 production under solar light. Rare earth elements have great potential in the application of solar energy conversion. Ce doping in the lattice of TiO 2 provides a new strategy for designing photocatalysts for photocatalytic water splitting.