A new Bi-based visible-light-sensitive photocatalyst BiLa1.4Ca0.6O4.2: crystal structure, optical property and photocatalytic activity

A new compound of BiLa1.4Ca0.6O4.2 is synthesized through solid state reaction, where the Ca substitutes, in part, the La site in a stable BiLa2O4.5 phase. The structure of the BiLa1.4Ca0.6O4.2 crystallizes in space group R3mH with a hexagonal lattice constants of a = 3.893(1) Å, c = 9.891(1) Å. Its optical absorption edge is about 2.05 eV, which just spans the visible light region. The photocatalytic activity of the BiLa1.4Ca0.6O4.2 powder to degradation of RhB under visible light irradiation is measured and improved more than 7 times by annealing in nitrogen ambient, indicating that annealing in nitrogen can effectively improve the photocatalytic activity by producing oxygen vacancy. Although the absolute photocatalytic activity obtained is low, there is great potential for enhancing the activity such as nanoscaling, doping, and coupling with other compounds.

, a compound possessing a monoclinic unit cell with superstructure, shows a yellow color suggesting its potential as visible light effective photocatalyst. However, our early experiments confirmed that its photocatalytic activity was poor compared with the above mentioned Bi-based oxides [13][14][15][16] .
To improve photocatalytic activity of a catalyst, various methods are used [23][24][25] , including the defect engineering 26 . Introduction of defects in a catalyst can boost interfacial charge transfer, thus effectively reducing recombination rate of photo-generated electrons-holes 26 . Pei et al. 27 revealed that higher defect concentration in ZnO enhanced photogenerated charge carriers separation and transportation, resulting in an enhanced photocatalytic activity. Kong et al. 28 revealed that tuning the defect concentration in TiO 2 leads to high photocatalytic activities. Many other methods like high temperature calcination 29 , cold plasma treatment 30 , annealing in atmosphere 31 , and particle bombardment 32 had been reported for creating defects in a catalyst. Compared with these methods, chemical doping is a simple and controllable way to adjust defects state and its concentration.
Herein, Ca is chosen to dope into BiLa 2 O 4.5 to modify its structure, properties and even its photocatalytic performance. A new Bi-based compound BiLa 1.4 Ca 0.6 O 4.2 is successfully synthesized. Its crystal structure, some optical properties and its photocatalytic activity are studied. It is found that the structure of BiLa 2 O 4.5 changes Scientific RepoRts | 6:23235 | DOI: 10.1038/srep23235 from monoclinic to rhombohedral by doping of Ca. The surface oxygen vacancy concentration of the sample is modulated by annealing in a nitrogen ambient. The photocatalytic activitie of sample is increased by annealing in nitrogen.

Results
In order to analyze the morphology and crystallite size, SEM was adopted for studying the BiLa 1.4 Ca 0.6 O 4.2 without and with annealing in a nitrogen ambient. From Fig. 1, it could be seen that there is no discernible change in the morphologies of samples with/without annealing in nitrogen ambient, and the average crystallite size is about 2 μm. Figure 2(A) shows the data of the X-ray powder diffraction pattern of the samples doping the BiLa 2 O 4.5 with different Ca content, and Fig. 2(B) is the partial enlarged drawing of Fig. 2(A). From Fig. 2(B), with the increase of Ca content, the peak at 27.5° gradually shifts to higher angle, and the peak at 31.5° gradually reduces and disappears. The structure of BiLa 2−x Ca x O 4.5−δ changes from monoclinic C2/m to rhombohedral R3mh with the increase of x value. A pure phase of BiLa 1.4 Ca 0.6 O 4.2 is obtained when x = 0.6 with a rhombohedral R3mh structure. The peak at 27.3° is emerged when x > 0.6, which means that the sample has emerged a new impure phase. The main diffraction peaks of BiLa 1.4 Ca 0.6 O 4.2 can be indexed using a rhombohedral cell with hexagonal lattice parameters a = 3.893(1) Å, c = 9.891(1) Å. The main diffraction peaks of BiLa 2 O 4.5 can be indexed using a monoclinic cell with lattice parameters a = 6.8272(6) Å, b = 3.9885(2) Å, c = 4.0523(5) Å, which is consistent with the literature 22,33 . The peaks marked by * belongs to superstructure [33][34][35][36][37] , and the hexagonal lattice parameters are a = 31.144 (8) Å, c = 19.782 (2) Å.
The structure of BiLa 1.4 Ca 0.6 O 4.2 is then refined through Rietveld refinement on the diffraction data. The simulated result is shown in Fig. 3. The final agreement factors converge to R p = 9.66%, R wp = 12.86%, and R exp = 3.29%, validating the reliability of our results. The refined lattice parameters are a = 3.893(1) Å, c = 9.891(1) Å for the hexagonal cell, which are consistent with that reported in the literature 33 . The inset of Fig. 3 shows the crystal structure deduced from the refinement, and the Bi, La and Ca atoms are randomly occupied the atomic sites marked with the bright blue balls.    Figure 5(A) shows the O 1s peaks of two types of the samples (i.e. as-prepared sample and the sample annealed in nitrogen). The O 1s state splits into two peaks, Ι and Π , located at 528.6 eV and 531.2 eV, respectively. The peak Ι can be attributed to the lattice oxygen in the BiLa 1.4 Ca 0.6 O 4.2 , while the peak Π can be due to the adsorbed oxygen in the surface of the sample, such as O − or OH − 38-40 . The strong peak Π in the two types of samples could be ascribed to that the samples are composed of the micrometer-sized BiLa 1.4 Ca 0.6 O 4.2 particles, whose specific surface area is high corresponding to a large bulk single crystal. The peak intensity ratio of Π /Ι are about 6.58, 6.95 for the as-prepared samples and the samples annealed in nitrogen ambient, respectively, indicating that the amount of adsorbed oxygen increases as more oxygen vacancies are created 41 .

Discussion
The density of states (DOS) of the valence band of samples was also measured by valence band XPS ( Fig. 5(C)). The valence bands (VB) are about 1.65 and 1.76 eV relative to the standard hydrogen electrode (SHE) for the samples prepared and annealed in nitrogen, respectively. Combined with the results from optical measurement, we can conclude that the conduction band (CB) minimum would occur at about − 0.4 and − 0.49 eV for the samples prepared and the annealed in nitrogen, respectively.
According to the UV-vis diffuse reflectance spectra and the valence-band XPS spectra, energy band edge diagram of the BiLa 1.4 Ca 0.6 O 4.2 is drawn as shown in Fig. 5(B). As illustrated in Fig. 5(B), the band edges of the CB      Fig. 6(A), the photocatalytic activities of samples are increased after annealing in nitrogen. The first order reaction kinetics of degradation of RhB under visible light is applied. The general pseudo-first-order model is shown as follows where C 0 and C are the concentrations of the RhB dye in solution at time 0 and t, respectively, and k is the pseudo-first-order rate constant 45 . Figure 6(B) displays photocatalytic reaction kinetics of degradation RhB according to the data plotted in Fig. 6(A). It can be seen from Fig. 6(B) the RhB degradation rates are about 0.00803 and 0.05719 h −1 for the as-prepared samples and the samples annealed in nitrogen ambient, respectively. This indicates that the sample annealed in nitrogen shows a more than 7 times improvement in photocatalytic activity over that of the as-prepared sample. Moreover, compared with the commercial P25, the photocatalytic activity of the prepared BiLa 1.4 Ca 0.6 O 4.2 annealed in nitrogen is significantly improved under visible light 46 . To better analyze the reasons for the improvement of the photocatalytic degradation rate, we measured the specific surface area (BET) of the samples by the static method. After annealing in nitrogen, the specific surface area increases from 2.39 m 2 /g to 6.74 m 2 /g which is improved by 2.82 times. Larger specific surface area make more photocatalytic reaction occur 46 . From Fig. 6(A,B), it is found that the photocatalytic activities of the sample annealed in nitrogen is increased compared with that of the as-prepared sample. For the samples annealed in nitrogen, the absorption edge is blue-shifted and the band gap is increased (Fig. 4(B)). Meanwhile, the surface oxygen vacancy concentration is also increased (Fig. 5(A)). The broaden band gap would lead to less visible light asbsorbed. But, the introduction of defects as oxygen vacancy can boost the interfacial charge transfer, thus effectively reduce recombination rate of photo-generated electrons-holes 26 . Therefore, a net effect is that the photocatalytic activities of samples annealed in nitrogen are highly enhanced.
In summary, a new compound of BiLa 1.4 Ca 0.6 O 4.2 is synthesized by solid state reaction, whose crystal structure is a rhombohedral R3mH structure with the hexagonal lattice constants of a = 3.893(1) Å, c = 9.891(1) Å. Its optical absorption edge is about 2.05 eV possessing an indirect band gap feature, which just spans the visible light region, validating it is a visible light sensitive photocatalyst. Its band gap is widened by oxygen vacancy produced from annealing in nitrogen. The photocatalytic activity of the BiLa 1.4 Ca 0.6 O 4.2 powder to degradation of RhB is measured and improved more than 7 times by annealing in nitrogen ambient. The improved photocatalytic activity is ascribed to the increased oxygen vacancy concentration on surface of sample while annealing in nitrogen ambient. The current results suggest that further modification of BiLa 1.4 Ca 0.6 O 4.2 by semiconductor coupling, noble metal decoration and nanocrystallization will doom improving its photocatalytic activity and providing new opportunities for solving the environmental pollution. Characterization. The crystal structure of BiLa 1.4 Ca 0.6 O 4.2 was analyzed by collecting powder X-Rar diffraction (XRD) patterns performed on a PANalytical Pert Pro X-ray diffractometer using CuKa radiation at room temperature. The morphology of the samples was analyzed by a scanning electron microscopy (SEM) with a model Hitachi S-4800. UV-vis diffuse reflectance spectra of the samples were collected on Shimadzu UV-3600 Plus to reveal their band structure information. The core level spectra of oxygen atoms in the samples were collected in an X-ray photoelectron spectroscopy (XPS) of Pekin Elmer PHI-5300 XPS instrument to study the bonding configurations of the oxygen atoms. The Brunauer-Emmett-Teller (BET) specific surface area was measured with a V-Sorb 2800 apparatus.

Photocatalytic activities.
The photocatalytic experiments were carried out in a home-made photochemical Reactor equipped with a 300 W Xe lamp. To acquire visible light, a 420 nm cut off filter was applied between the lamp and the catalyst container. Rhodamine B (RhB) was adopted as the photocatalysis probe due to its high stability and sensitivity to visible light absorption due to its intrinsic absorption band at about 553 nm. In each photocatalytic experiment, 50 ml RhB aqueous solution in concentration 10 mg/L was filled into quartz beaker together with 50 mg BiLa 1.4 Ca 0.6 O 4.2 powder. Before irradiation, the mixture was magnetically stirred in dark for 1.5 hours to approach adsorption-desorption equilibrium. The mixtures were then irradiated by visible light at room temperature and ambient pressure, while stirring to keep catalyst particles homogenously dispersed in solution.
During visible light irradiation, about 5 mL of the suspension was taken out from the beaker at a given time intervals about 2 hours in sequence for subsequent analysis of target dye concentration after centrifuging. Absorption spectra of the suspensions were collected by a Shimadzu UV-3600 Plus spectrometer. The photocatalytic activity of the BiLa 1.4 Ca 0.6 O 4.2 was evaluated from the intrinsic absorption band (at 553 nm) intensity ratio of the remnant RhB after visible light illumination to that of the RhB in parent solution.