High-performance visible-light active Sr-doped porous LaFeO3 semiconductor prepared via sol–gel method

ABSTRACT In this work, we have successfully fabricated Sr-doped porous LaFeO3 samples via sol–gel method. The results reveal that Sr2+ cation is effectively doped into LaFeO3 crystal lattice substituting La3+ cation. The visible light catalytic performance of the materials was evaluated by the degradation of 2,4-dichlorophenol (2,4-DCP) and Rhodamine B (RhB). The amount-optimized Sr-doped porous LaFeO3 sample exhibited outstanding visible-light catalytic performance for the degradation of the model pollutants compared to the porous LaFeO3 alone. The enhanced performance was accredited to the enlarged surface area, absorption extension via the surface states of the introduced Sr2+ below the conduction band bottom of LaFeO3, and promoted charge carrier’s separation as confirmed by various experiments. Experiments of radical trapping reveal that •OH species are dominant intermediate oxidants involved in the oxidation of 2,4-DCP and RhB over the optimized sample. This research will provide new routes for environmental remediation based on the LaFeO3 semiconductor.


Introduction
Water pollution has disturbed the equilibrium of melodious co-existence between living organisms and nature (1,2). In daily life, the toxic organic/inorganic byproducts are released in industrial waste water and house sewage that could be efficiently degraded to protect the natural ecosystem (3,4). The most common and environmental hazardous pollutants include 2,4-DCP and RhB dye. These pollutants are highly toxic and carcinogenic, therefore enlisted by the United States Environmental Protection Agency (U.S. EPA) as a high-risk pollutant to be controlled (5,6). The commonly employed methods do not fulfill the requirement of efficient degradation of these pollutants (7). Thus it is necessary to replace the conventional techniques by advanced oxidation techniques.
Semiconductor photocatalytic technique has gained marvelous attention in the present decade due to its abundant virtues such as cost effective, non-toxicity, high chemical and thermal stability, and environmentally friendly nature (8,9). Different wide band gap semiconductors such as TiO 2 (10) ZnO (11), SnO 2 (12), and so on, have been widely used in photocatalysis for pollutants degradation due to their lower price, chemical/thermal stability, appropriate valence and conduction band potentials and low environmental impact. These semiconductors usually exhibit poor performance for pollutants degradation due to the limited solar energy consumption, i.e. ca. 4% (only UV range). To accomplish high performance for pollutants degradation, the use of narrow band gap semiconductors is highly recommended (13). The narrow band gap photocatalysts could absorb a wide range of solar spectrum that mainly comprises the visible light region.
Among the narrow band gap semiconductors, Lanthanum Ferrite (LaFeO 3 ) with general formula ABO 3 has received remarkable interest in photocatalysis due to its unique electronic structure (14). In LaFeO 3 oxide, La belongs to rare-earth elements while Fe belongs to the 3d transition metals. The presence of Fe element is very important because the catalytic processes depend upon redox behavior of Fe cation. Further, LaFeO 3 can probably utilize visible light of solar spectrum because its band gap is only 2.0 eV (15). Nevertheless, the pollutant degradation performance of LaFeO 3 is still poor owing to the small specific surface area, short carrier diffusion lengths and the rapid recombination rate of charges (16).
To promote the photocatalytic performance of LaFeO 3 , various modification strategies such as nanostructured fabrication (17), doping elements (18), semiconductors coupling (19) and creating pores (20) are widely employed. In various modification strategies, creating pores and doping elements are more beneficial because porous materials could provide more surface area for catalytic reactions and doping could extend the visible light response of nanomaterials. For example, Zhang et al. (21) fabricated mesoporous LaFeO 3 that exhibited enhanced catalytic activity for methyl chloride (CH 3 Cl) combustion, which was accredited to the extremely large specific surface area due to porous morphology. In another work, Phan et al. (22) fabricated Cu-doped LaFeO 3 catalyst that showed high performance for decolorization of cation and anion dyes under visible light irradiation.
In present work, we have fabricated Sr-doped porous LaFeO 3 samples and explored their photocatalytic activities for organic pollutants degradation. So far, there have been no previous reports on the model pollutants, i.e. 2,4-DCP and RhB degradation over the Sr-doped porous LaFeO 3 . Thus this work has a great scientific value in the field of photocatalytic environmental remediation. According to the experimental results, the optimized 1.5Sr-LaFeO 3 sample exhibited excellent catalytic performance for 2,4-DCP and RhB degradation under visible light. This research will provide new routes for the development of highly proficient visible light active LaFeO 3 -based catalysts for environmental decontamination.

Chemicals
Analytical grade high purity chemical reagents were purchased and directly used in this research. The solvent deionized water was utilized throughout the experimentation.

Synthesis of porous LaFeO 3 and Sr-LaFeO 3 photocatalysts
To prepare porous LaFeO 3 , an equimolar (0.02 mol) amount of Fe(NO 3 ) 3 ·9H 2 O and La(NO 3 )·6H 2 O precursors was dissolved in a water/ethylene glycol (EG) solvent mixture (50/50 vol %) at room temperature. The solution was kept in ultrasonic bath for 30 min and then 100 nm size polystyrene (PS) spheres (size = 200 nm) were drenched in the solution under stirring for 6 h. The beaker-containing solution was put into the water bath and heated at 120°C to get a gel, which was then dried at 85°C in an oven and annealed at 600°C (temp ramp 1°C min −1 ) to get porous LaFeO 3 . To prepare Sr-doped porous LaFeO 3 , the same procedure was used except for Sr(NO 3 ) 3 ·H 2 O precursor was used as a source of Sr. Porous LaFeO 3 samples containing different mass percentages of Sr (i.e. x = 0.5, 1, 1.5, 2%) were prepared.

Characterization of materials
The XRD pattern of the catalysts was determined via the Rigaku-D/MAX-rA (XRD, made in Japan) powder diffractometer. Shimadzu UV-2550 Spectrophotometer was used for recording UV-visible diffuse reflectance spectra. The Fourier Transform Infrared (FTIR) spectra were measured with Thermo-Scientific Nicolet-IS10 spectrometer equipped with ATR. JEOL-JEM-3010 electron microscope (made in Japan) was used for capturing TEM/HRTEM images. The BET isotherm curves were obtained with Micro-meritics ASAP-2020 M system. ST-2000 adsorption instrument was used BET surface area evaluation. The energy dispersive X-ray (EDX) spectra were obtained with the TEM-EDX. The XPS spectra were investigated using Al (mono) X-ray source with Kratos-Axis Ultra DLD. A PE-LS-55 spectro-fluoro-photometer with wavelength of 325 nm was used to obtain the photoluminescence (PL) spectra. Electrochemical impedance spectroscopy (EIS) spectra were measured via the Princeton Applied-Research, Versa, STAT-3 ( f = 10 −2 -10 5 Hz) and 10 mV (RMS) amplitude was employed at 0.4 V potential vs. the Ag/AgCl electrode in Na 2 SO 4 electrolyte (0.5 M L −1 ). A high-power Xe lamp (300 W) with 420 nm cut-off wavelength was employed as a source of light. Electron paramagnetic resonance (EPR) spectra were measured with a Bruker D-200 instrument (IBM-Bruker) at room temperature.

Films preparation and photoelectrochemical analysis
For films preparation, following procedure was used. About, 20 mg of the sample powder was highly dispersed in 1 mL IPA (isopropyl alcohol) through ultrasonication for 20 min and then vigorously stirred for another 20 min. Then, 20 mg of polymer (Macrogol-(6000)) was highly grinded into fine powder and added to the dispersion. The dispersion was ultrasonically treated for 20 min and then stirred for another 20 min. Further, an appropriate volume of acetyl acetone (i.e. 0.05 mL) was added dropwise and kept under stirring for 72 h. After that, the paste was dropped onto the conductive side of a well-cleaned fluorine doped tin-oxide (FTO) glass already calcined at 400°C to prepare film. The film was dried and calcined at 500°C for 20 min. Finally, the FTO glass with film was cut into parts (i.e. 1.0 cm × 3.0 cm) having film surface area of 1.0 cm × 1.0 cm.

Pollutant degradation activity evaluation
A quartz cell was used for 2,4-DCP degradation activities. First, solution of 2,4-DCP (i.e. 10 mg L −1 concentration) was prepared in a 1-L volume flask. For each experiment, a 60-mL solution of 2,4-DCP was taken in quartz cell and 100 mg catalyst powder was added to it. After stirring for 0.5 h in dark (i.e. adsorption/desorption equilibrium), the reactor was irradiated for 2 h under high power (300 W) Xe-lamp with 420 nm wavelength cut-off filter. Then, a desired amount was centrifuged and transferred to a quartz cell. The 2,4-DCP concentration was evaluated at 285 nm absorption wavelength with the help of Perkin-Elmer Lambda-35 spectrometer (USA). The RhB degradation experiments were performed via the following procedure; 100 mg of catalyst sample was dispersed in 50 mL solution of RhB dye (i.e. 10 mg L −1 ) and stirred for 30 min in dark (i.e. to accomplish adsorption equilibrium). Subsequently, a high power (300 W) Xe lamp with 420 nm cut-off filter was employed as an irradiation source and the reactor was kept under stirring and irradiated for 2 h. After a specific interval of time, the RhB concentration was analyzed at wavelength 553 nm, by taking appropriate amount of solution through a syringe, centrifuging and analyzing via the Shimadzu UV-2550 photometer.

Evaluation of hydroxyl radical (•OH)
50 mg catalyst powder was dispersed in 40 mL solution of coumarin (i.e. 10 mg L −1 ) contained in a quartz cell to analyze hydroxyl radical amount. A high power (300 W) Xe-lamp with 420 nm cut-off lens was employed as irradiation source for 1 h under stirring. A proper volume was transferred to a Pyrex glass-cell and the hydroxyl radical amount was detected at 350 nm excitation wavelength via the coumarin fluorescence method.  (23). For Sr-doped LaFeO 3 samples, an obvious shift toward larger diffraction angle was observed as can be seen from Figure 1(B). It is demonstrated that the ionic radius of Sr 2+ (1.44 Å) is slightly larger than that of La 3 + . Hence, the substitution of La 3+ ion with Sr 2+ results in the lattice distortion of LaFeO 3 (24). This proves that Sr cations are successfully doped into the porous LaFeO 3 crystal and substituted La cations. Figure 1(C) shows the UV-Visible absorption spectra of the catalysts. To estimate the energy bang gaps of porous LaFeO 3 and Sr-doped LaFeO 3 samples, Tauc plots were determined using Kubelka-Munk method (25). For porous LaFeO 3 sample, the estimated band gap was 2.09 eV ( Figure  1D). The estimated band gaps for 0.5Sr-LaFeO 3 , 1Sr-LaFeO 3 , 1.5Sr-LaFeO 3 , and 2Sr-LaFeO 3 samples were 2.06, 2.0, 1.91 and 1.79 eV, respectively. This clarifies that Sr doping has effectively reduced the band gap of porous LaFeO 3 . Figure 2(A) shows the TEM image of porous LaFeO 3 . As can be seen that porous LaFeO 3 exhibit crystallite size of around 100 nm and the pores can be observed. The selected area HRTEM image ( Figure 2B) demonstrates that the fringes with dspacing of 0.28 nm are attributed to LaFeO 3 . The TEM image of 1.5Sr-LaFeO 3 sample ( Figure 2C) reveals that the morphology of porous LaFeO 3 is slightly changed due to Sr-doping. This might be due to the lattice distortion of porous LaFeO 3 . The lattice fringes can be seen in the selected area HRTEM image of 1.5Sr-LaFeO 3 sample ( Figure 2D). The EDX spectrum of porous LaFeO 3 ( Figure  2E) shows the presence of La, Fe and O element peaks, while that of the 1.5Sr-LaFeO 3 sample ( Figure 2F) shows Sr, La, Fe and O element peaks. This shows the successful doping of Sr cation into the crystal lattice of porous LaFeO 3 . The N 2 adsorption and desorption isotherm curves and the resultant pore size distribution curves of porous LaFeO 3 and 1.5Sr-LaFeO 3 samples are revealed in Figure 3. As clear, the sorption isotherm curves of LaFeO 3 and 1.5Sr-LaFeO 3 samples ( Figure 3A) demonstrate hysteresis loops, which is the characteristic of porous materials (26). The Brunauer-Emmett-Teller (BET) surface area of porous LaFeO 3 was 28.56 m 2 g −1 .

Structural morphology and composition
The average pore size of LaFeO 3 was about 14.56 nm ( Figure 3B). The BET-specific surface area and average pore diameter of 1.5Sr-LaFeO 3 sample was 38.77 m 2 g −1 and 12.64 nm, respectively. The decrease in pore diameter of 1.5Sr-LaFeO 3 may be due to the partial loss of structural order of porous LaFeO 3 . To confirm the surface composition, Fourier Transform Infrared FTIR spectra of LaFeO 3 and the optimized 1.5Sr-LaFeO 3 samples were measured in the range 400-4000 cm −1 as revealed in Figure S1.  Figure 4A). The La3d high-resolution XPS spectra of porous LaFeO 3 and 1.5Sr-LaFeO 3 samples are shown in Figure 4(B). The binding energies of La3d 5/2 and La3d 3/2 orbitals are located at 833.83 and 850.74 eV, respectively, corresponding to the La cation 3 + oxidation state (27). As obvious, the binding energy peak of 1.5Sr-LaFeO 3 sample is slightly shifted toward lower energy side. This confirms the partial substitution of La cations by Sr cations. The high-resolution Fe2p XPS spectra of porous LaFeO 3 and 1.5Sr-LaFeO 3 samples are revealed in Figure 4(C). The binding energy peaks of Fe2p 3/2 and Fe2p 3/2 orbitals are respectively located at 710.6 and 723.6 eV, corresponding to the Fe +3 oxidation state in porous LaFeO 3 (28). The binding energy peaks of 1.5Sr-LaFeO 3 sample are also slightly blue-shifted. The high-resolution O1s XPS spectra of porous LaFeO 3 and 1.5Sr-LaFeO 3 samples are shown in Figure 4

Charge separation
Photoluminescence (PL) is a sensitive technique used to examine the structure and properties of active sites on materials surfaces. Thus the PL technique is very helpful for understanding the surface phenomenon in semiconductor photocatalysis. In addition, the PL technique gives us evidence of the defects, oxygen vacancies, charge trapping, immigration and transfer. It is assumed that the PL response could mainly derive from the electronic transition, occurring from conduction bands to the valence bands of semiconductors.
Higher the PL intensity of nanomaterials, charge recombination would be high, and vice versa (30). As obvious from Figure 5(A), the PL intensity of porous LaFeO 3 is very high. Principally, the PL intensity of Sr-doped LaFeO 3 samples is remarkably decreased and the lowest intensity was observed for the 1.5Sr-LaFeO 3 sample. This implies that charge recombination in the 1.5Sr-LaFeO 3 sample is drastically reduced. Generally, the photocurrent response of semiconductors could reveal an indication to the catalyst capability to produce and transfer the photo-induced charges indirectly, which greatly impacts the photocatalytic performance of catalysts. The photocurrent I-V curves of porous LaFeO 3 and Sr-doped porous LaFeO 3 electrodes under visible light are shown in Figure 5(B). Porous LaFeO 3 exhibited weak photocurrent density response, while for Sr-doped samples, it is quite obvious. The photocurrent density response of 1.5Sr-LaFeO 3 sample was remarkably improved suggesting the superior charge separation. The photocurrent I-t plots of the porous LaFeO 3 and Sr-doped porous LaFeO 3 electrodes under visible light are shown in Figure 5(C). As obvious, the porous LaFeO 3 showed a weak photocurrent response. However, enhancement in photocurrent was observed for Sr-doped porous LaFeO 3 samples. The highest photocurrent was observed for 1.5Sr-LaFeO 3 sample, which indicates fast charge transfer and highly efficient charge carriers separation through the 1.5Sr-LaFeO 3 electrode interface. This could drastically enhance the photo-activity for pollutants degradation. The effective charge separation in Sr-doped porous LaFeO 3 samples can be further verified by Nyquist plots. As revealed in Figure 5(D), the Nyquist arc radius of porous LaFeO 3 sample is quite large and it is strongly decreased in case of Sr-doped LaFeO 3 samples. The lowest arc radius was observed for 1.5Sr-LaFeO 3 sample, which further supports the suppression of charge carrier's recombination. The PEC and EIS results well support the PL ones.

Visible light activities
The catalysts performance was appraised for photocatalytic degradation of 2,4-DCP for 2 h irradiation under visible light. The characteristic absorption spectra of 2,4-DCP and RhB after photocatalytic reaction in the presence of porous LaFeO 3 and Sr-doped porous LaFeO 3 samples were measured as provided in Figure  S2 A and B, respectively. As clear from Figure   the optimized 1.5Sr-LaFeO 3 sample degraded 68% of 2,4-DCP pollutant. The catalysts performance was further appraised for Rhodamine B (RhB) degradation as shown in Figure 6(B). After stirring in dark for 30 min (i.e. adsorption equilibrium), the glass reactor was irradiated with visible light source and the desired volume of solution was collected at regular time interval (i.e. 30 min) for measuring the RhB concentration through UV-Visible spectrophotometer at 553 nm wavelength. The RhB degradation over Sr-doped LaFeO 3 samples is remarkably high compared to that of the porous LaFeO 3 sample. The highest activity was detected for 1.5Sr-LaFeO 3 sample (i.e. 71%) which fully supports the PEC, EIS and PL results. Thus, the improved photoactivities might be resulted from the enlarged specific-surface area due to porous morphology, enhanced light absorption and improved charge separation via doping Sr cations. The performance of our photocatalyst for pollutants degradation is better compared to those of the previous reports as mentioned in Table S1. Generally, it is believed that •OH is one of the most active intermediates involved various reactions with inorganic and organic molecules (31). To investigate the •OH formation, coumarin-fluorescence technique was employed to identify the •OH species amount produced by each sample during photocatalytic process. Figure 6(C) reveals that porous LaFeO 3 sample produced a small quantity of •OH as confirmed by the weak fluorescence intensity. However, the Sr-doped porous LaFeO 3 samples produced high quantity of •OH. Notably, the optimized 1.5Sr-LaFeO 3 sample produced significant quantity of •OH, which further suggests enhanced charge separation. In photocatalysis, it is important to recognize the main oxidant species involved in the degradation of pollutants. In aqueous media, the photoinduced holes (h + ), superoxide radical species (O 2 •− ) and hydroxyl radical species (•OH) are important intermediates concerned with the degradation of organic pollutants. Commonly, ethylenediaminetetraacetic acid disodium salt (EDTA-2Na), para-Benzoquinone (BQ), and isopropyl alcohol (IPA) scavengers are used as trapping agents to capture the photo-generated h + , O 2 •− and •OH during photocatalytic processes (32,33). These scavengers were used to prove which specie is governing in the oxidation of 2,4-DCP and RhB pollutants. The experiments were completed under the same conditions with the aid of scavengers. As can be seen in Figure 6(D), without adding scavengers, the 2,4-DCP and RhB degradation over the optimized 1.5Sr-LaFeO 3 sample was 68% and 72%, respectively, under same experimental conditions. Upon the addition of few drops of 1 mmol scavengers' solution, the 2,4-DCP and RhB degradation was decreased up to some extent in the existence of EDTA-2Na and BQ scavengers. Interestingly, it was strongly inhibited in the aid of IPA scavenger. These results demonstrate that •OH species are involved as dominant intermediates in the degradation process over the 1.5Sr-LaFeO 3 sample. To further confirm the scavenger's trapping experiments, electron paramagnetic resonance (EPR) analysis was carried out to detect the reactive intermediates during photocatalysis. The trapping agent 5,5-Dimethyl-L-pyrroline N-oxide was employed to trap the active intermediates. As clear from Figure S3,  The stability of the sample is an important parameter for evaluating its photocatalytic performance. Therefore, the stability of 1.5Sr-LaFeO 3 sample was appraised for degradation of 2,4-DCP and RhB, by repeating the reaction 04-times (each of 3 h). As clear from Figure 6(E, F), the degradation of 2,4-DCP and RhB over the 1.5Sr-LaFeO 3 sample changes little, and even 66% and 70% of 2,4-DCP and RhB pollutants were degraded after recycling four times. This clarifies that the 1.5Sr-LaFeO 3 sample exhibit high stability during photocatalytic processes. To further confirm the stability of the optimized 1.5Sr-LaFeO 3 catalyst, the XRD and SEM analysis was carried out as provided in Figures S4 and S5. The structure and morphology of the catalyst before and after the catalytic reaction is almost the same. This further confirms that the as-prepared 0.15Sr-LaFeO 3 photocatalyst is highly stable.

Mechanism
According to the Tauc's plot of Kubelka-Munk method, the band gaps of porous LaFeO 3 and 1.5Sr-LaFeO 3 catalysts were respectively predicted to be 2.09 and 1.91 eV. Therefore, it is important to investigate the mechanism of pollutants degradation over 1.5Sr-LaFeO 3 sample under visible light irradiation. To confirm that Sr dopant has introduced surface states near conduction or valence band of porous LaFeO 3 , XPS valence band spectra of porous LaFeO 3 and 1.5Sr-LaFeO 3 samples were measured. As obvious from Figure S6, the valence bands of the both samples were located at 2.4 V, which means that Sr dopant introduced surface states exists close to the bottom of the conduction band of porous LaFeO 3 . Thus, the Sr introduced new energy levels near the bottom of the conduction band of porous LaFeO 3 has notably reduced its band gap to 1.91 eV. As the valence band potential and band gap energy values are known, thus the conduction band values can be calculated by using the equations as follows (34,35): Herein, E VB denotes valence band potential value, E CB denotes conduction band potential value, χ denotes the absolute electronegativity, (where, a, b, c, d denote atomic number of the compounds), E g represents band gap energy of the catalysts, and E e is the free electron energy vs the H 2 electrode (4.5 V). Thus the conduction band values of porous LaFeO 3 and 1.5Sr-LaFeO 3 catalysts were predicted to be 0.4 and 0.49 V, respectively, as depicted in Figure 7. Thus the visible light irradiation upon interaction with the surface of 1.5Sr-LaFeO 3 catalyst will initiate the photo-catalytic reactions by absorbing light energy (hv) equal to, or greater than its respective band gap energy. This will lead to the generation of charge carriers. The excited electron will transfer from the valence band of 1.5Sr-LaFeO 3 to its conduction band, meanwhile the holes will remain in its valence band. The photo-induced holes in its valence band would be trapped by the surface hydroxyl groups (OH -) of the photocatalyst and generate •OH, which are very efficient in pollutant degradation. The excited electrons in the CB would react with the adsorbed surface oxygen and produce superoxide anion radicals. The photoinduced valence band holes also directly contribute to the degradation. The scavengers-trapping experiments revealed that •OH species are the dominant reaction intermediate involved in pollutant degradation. From the experimental results, it is concluded that the activity enhancement of the catalysts is related to the enlarged surface area due to porous morphology and band gap narrowing due to the Sr introduced surface states in porous LaFeO 3 .

Conclusion
In this work, Sr-doped porous LaFeO 3 samples are prepared via sol-gel method. The results reveal that Sr 2+ cation is successfully doped into the lattice of LaFeO 3 and substituted La 3+ cation. The optimized 1.5Sr-LaFeO 3 sample exhibited excellent visible light catalytic activities for RhB and 2,4-DCP degradation compared to the porous LaFeO 3 . Radical trapping experiments reveal that •OH species are the dominant reactive intermediates involved in the degradation of RhB and 2,4-DCP pollutant over the optimized 1.5Sr-LaFeO 3 sample. The newly designed work will open new directions for the fabrication of LaFeO 3 -based high-performance catalysts for environmental remediation.