Abstract
The shape and connection type of MOx are critical to the physical and chemical properties. A series of new material Sr2-xNaxNbO4 containing NbO4 tetrahedra was prepared by controlling the ratio of SrCO3 to sodium niobate under ambient air. With increasing the content of Sr in the sample, the MOx shape will change from NbO6 octahedra to NbO4 tetrahedra, which is confirmed by the Raman scattering spectra. With increasing the content of NbO4 in the sample, the lattice parameter increases, optical band gap becomes larger, and the surface changes to be more active for oxygen adsorption, resulting in a higher photocatalytic activity.
1. Introduction
Complex metal oxide (AMOy) is built upon the framework of MOx with sharing edges or corners and the inserted metal ion. The shape of MOx is diverse, such as octahedron and tetrahedron. The shape and connection types of MOx are critical to the physical and chemical properties [1–5]. Recently, theoretical predictions and experimental results confirmed that the TaS2 built from the edge-sharing of TaS6 trigonal prisms exhibits superior catalytic performance in comparison with that of TaS6 octahedra [6, 7]. Those early studies suggested a potential way to improve the catalytic activity via tuning the shape of MOx. The above hypothesis is rarely investigated in niobates, which is one kind of photocatalyst for water splitting and organic degradation, such as NaNbO3, SrNb2O6, and Cs2Nb4O11 [8–15]. Most of the niobate photocatalysts contain NbO6 octahedra. For instance, NaNbO3 contains the formwork of NbO6 octahedra sharing corners with a slight distortion [16, 17]. The NbO6 octahedra in SrNb2O6 share edges and corners simultaneously [14]. Cs2Nb4O11 is the one that coexisted with NbO6 octahedra and NbO4 tetrahedra in the structure [12, 18]. The tetrahedral NbO4 structure is rarely found in niobium oxide compounds because the Nb5+ atom is usually too large to fit into an oxygen-anion tetrahedron and exists in the rare earth ANbO4 (A = Y, Sm, and La) compounds [1, 2, 19]. To investigate the effect of MOx shape on photocatalytic activity, the challenge is preparing the sample containing NbO4 tetrahedra.
Here, we reported a series of new material Sr2-xNaxNbO4 containing NbO4 tetrahedra, which was prepared from sodium niobate and SrCO3 with different ratios under ambient air. The MOx shape will change from NbO6 octahedra to NbO4 tetrahedra. The Raman results identified that NbO4 tetrahedra existed in the samples. The change from the NbO6 octahedra to NbO4 tetrahedra results in the different physical and chemical properties, such as optical band gap, adsorption property, and photocatalytic activity.
2. Experimental Section
2.1. Sample Preparation
Sodium niobate precursor was synthesized via a simple hydrothermal process. Typically, the prepared mixture of 50 mL of NaOH solution (8 M) and 2 g of Nb2O5 powder was poured into a 100 mL teflon-lined stainless steel autoclave. The hydrothermal reaction was performed in a drying oven at 120°C for 3 h [20–22]. After naturally cooling down to room temperature, we obtained the samples and rinsed the samples with deionized water and absolute ethanol to remove the residual unreacted precursor. The sodium niobate precursor was confirmed as the Na2Nb2O6•xH2O from the XRD.
SrCO3 and sodium niobate precursor were mixed in the mortar at different molar ratios and grounded for 15 min. Then, they were heated at 500°C for 3 h and calcined at 800°C for 10 h in a muffle furnace. After cooling, we got the samples. For simplicity, we denoted the samples prepared from the ratios of SrCO3 to sodium niobate precursor of 1 : 2, 2 : 2, and 4 : 2 as A, B, and C, respectively. For comparison, the sample was also prepared by solid state reaction method using NaNbO3 and SrCO3 as starting materials at 900°C for 10 h.
2.2. Characterizations
The crystal structures were identified by X-ray diffraction (XRD). Raman scattering spectrum was measured using a laser Raman spectrophotometer at room temperature. The X-ray photoelectron spectra (XPS) were measured by a Kratos AXIS Ultra photoelectron spectroscope. The diffusion reflection spectrum was recorded with a UV-vis spectrophotometer (Shimadzu 2550) using BaSO4 as the reference and transformed to the absorption spectra automatically. The temperature programmed desorption (TPD) was carried out on Chembet Pulsar TPD using He pretreatment at 300°C for 1 h. Inductively coupled plasma (ICP) was performed on Agilent 720/730 after pretreatment in mixture of HCl and HNO3.
Photocatalytic activity for the decomposition of RhB in an aqueous solution was evaluated in the presence of samples A, B, and C under full arc light irradiation of Xe lamp as reported previously [23]. The initial concentration and pH value of RhB solution were about 2.5 mg L−1 and 4.5. Before illumination, the reagent was left in the dark for 30 min to achieve adsorption-desorption equilibrium. After adsorption of dye, the pH value is 6.7 for sample A, 6.6 for sample B, and 8.5 for sample C. The light emitted from the 300-W Xe lamp (the spectra is the same as that in our previous report [24]) directly irradiated on the solution in the reactor. The stirring was on during the reaction. The variation in the concentration of RhB was recorded by measuring the absorbance of the main peak in the UV-vis spectrum (UV-2550, Shimadzu) every 30 min.
3. Results and Discussions
All the samples show the similar XRD patterns with that of Sr2NbO4 (PDF#28-1245), as shown in Figure 1(a) [25]. A slight amount of SrCO3 impurity was found in sample C due to too much SrCO3 in the starting mixture. The compositions of samples were determined as Sr0.6Na1.4NbO4, Sr0.97Na1.03NbO4, and Sr1.73Na0.27NbO4 for A, B, and C, from the ICP results. The total charge at A site will increase from 2.8 for sample A, 2.97 for sample B, and up to 3.73 for sample C. Moreover, the positions of diffraction peaks were successively shifted towards smaller 2θ with increasing the content of SrCO3, as shown in Figure 1(b). Two factors, namely strains and lattice parameter changes, will contribute to the shift of diffraction peaks [26]. The strains could be estimated by the following equations. where is the full width at half-maximum (FWHM) of the θ-2θ peak; is the diffraction angle; is the X-ray wavelength; is the effective strain, and is the crystallite size. The strain () is calculated from the slope, and the crystallite size () is calculated from the intercept of a plot of against . The calculated effective strain of samples A, B, and C are 1.15%, 0.87%, and 0.41%, respectively. However, the decrease in the effective strain is not reasonable. Kumar et al. found that the diffraction peak of NaNbO3/CdS core/shell particles shifted to larger 2θ in comparison with pristine NaNbO3, where the effective strain of NaNbO3/CdS core/shell particles will decrease down to −0.76% from 0.89% for NaNbO3 [26]. Our result is opposite to the above tendency, so we thought the shift of diffraction peak is not caused by the strains. According to the Bragg’s law, we calculated the lattice parameter using Sr2NbO4 as the parent structure, as shown in Figure 1(c). It is clearly seen that the lattice parameter increases with the rise in the content of Sr and trend towards 4.11 Ǻ that of Sr2NbO4.
(a)
(b)
(c)
Raman scattering is an effective method of investigating the changes at local structure. Figure 2(a) shows Raman scattering spectra of samples prepared from different ratios. The Raman band near 780–840 cm−1 appears in all the samples, whereas the band near 500–600 cm−1 only existed in sample A. The Raman band around 780–830 cm−1 is corresponding to the vibrational modes of a regular NbO4 tetrahedron, and the band near 500–700 cm−1 is assigned as the symmetric stretching mode of the NbO6 octahedra [1, 18, 19]. The correlation between the Raman wavenumbers for the stretching bands in niobate compounds and the Nb-O bond length has previously been established. We calculated the Nb-O bond length using the empirical equation , where is the Nb-O stretching wavenumber, and is the Nb-O bond length [13, 27]. The results are shown in Figure 2(b), and all of the Nb-O length satisfied the requirement of Nb-O bond length in NbO4 tetrahedron of 1.83–1.93 Ǻ, indicating that all the samples contain NbO4 tetrahedron [1]. The precursor of sodium niobate only has the band near 500–600 cm−1, and the YNbO4 contains the Raman band near 780-840 cm−1. Above results confirmed that sample A contains NbO6 octahedran and NbO4 tetrahedra, and samples B and C mainly contain NbO4 tetrahedron. The formation mechanism will be discussed later. The Nb-O bond length increases with the rise in the content of Sr.
(a)
(b)
The optical properties of samples are displayed in Figure 3. The absorption edge of sample A locates at ~324 nm corresponding to ~3.8 eV, whereas that of samples B and C is ~310 nm corresponding to ~4.0 eV, implying that sample A has a narrower band gap in comparison with samples B and C. The increase of lattice distance usually results in the larger band gap due to the smaller expansion of the energy levels according to the principle of solid state physics. Consequently, it is reasonable to get the bigger optical band gap in samples B and C because they have larger lattice parameter. Samples B and C have the similar absorption edge and band gap, which is possibly attributed to their common properties of containing NbO4 tetrahedra. Sample C has a tail around 310–500 nm, indicating that sample C possibly has more defects in comparison with sample B.
The O1s XPS lines obtained from samples are shown in Figure 4. The binding energies were calibrated to the C1s peak at 284.8 eV. The O1s spectra can be divided into two peaks at 530.0 and 531.4 eV, respectively. The peak at 530.0 eV represents the binding energy of lattice oxygen ions [28–31]. The peak at 531.4 eV was considered as the binding energy of oxygen defects [31] or surface oxygen, such as O2− [29, 30]. The ratio of Oads to Olatt increases with the rise in the content of Sr. We used the TPD to investigate the property of oxygen adsorption. The amount of adsorbed oxygen on samples is consistent with the XPS results.
(a)
(b)
Photocatalytic activity was evaluated by the photocatalytic RhB photodegradation under 300-W Xe lamp. Figure 5(a) shows the RhB concentration varied with reaction time. Obviously, after irradiation, the RhB concentration is decreased. When the reagent solution is dilute, the reaction rate () can be expressed as , where is the apparent rate constant, and is the instantaneous concentration of the reactant. The plots of −ln() vs. time are shown in Figure 5(b), and the rate constants of RhB degradation are also displayed in Figure 5(c). is 2.0 × 10−2, 5.2 × 10−2, and 2.8 × 10−2 h−1 for samples A, B, and C, respectively. Sample B exhibits 2.6 times higher activity than sample A. Dichlorid phenol (DCP), a colorless toxic pollutant, was used as a model pollutant, as shown in Figure 5(d). The DCP concentration was decreased over the sample B.
(a)
(b)
(c)
(d)
We will address the formation mechanism of NbO4 tetrahedra and how to affect the photocatalytic activity. Usually the Sr2NbO4 could be synthesized under vacuum or reduced atmosphere [25]. In this work, we used sodium niobate as one of the starting materials and got Sr2-xNaxNbO4 under ambient air. When sodium niobate reacts with SrCO3, the Sr2+ will substitute Na+. The Sr atom will give one excess charge and create an equivalent reduced atmosphere, resulting in the change of NbO6 octahedra to NbO4 tetrahedra. With increasing the content of Sr in the structure, the lattice parameter will be increased due to larger ionic radius of Sr2+ (1.18 Ǻ) than Na+ (1.02 Ǻ) [16]. More amount of Sr inserted into the structure will cause the unbalance charge in A site (AMOx), leading to more defects generated in the sample, as that observed in sample C. Moreover, we could get the Na-doped Sr2NbO4 via the solid state reaction method using pseudoperovskite NaNbO3 as the starting material, indicating that Na is very important to get Sr2-xNaxNbO4, not the structure of sodium niobate. With increasing the content of NbO4 in the sample, supported by the results of Raman spectra, the lattice parameter increases, optical band gap becomes larger, and the surface changes to be more active for oxygen adsorption, resulting in a higher photocatalytic activity. Above trends could be obtained from the results of samples A and B, supporting the catalytic property of MO4 tetrahedron is superior to that of MO6 octahedron. Due to more defects in sample C, its photocatalytic activity decreases in comparison with sample B.
4. Conclusions
We prepared a series of Sr2-xNaxNbO4 photocatalytic materials containing NbO4 tetrahedra by controlling the ratio of SrCO3 to sodium niobate under ambient air. With increasing the content of NbO4 in the sample, the lattice parameter increases, optical band gap becomes larger, and the surface changes to be more active for oxygen adsorption, resulting in a higher photocatalytic activity. The efficiency of this catalyst can also be due to the photosensibilisation of RhB and should be tested by using another organic molecule as model pollutant.
Data Availability
The data used to support the findings of this study are available from the corresponding author upon request.
Conflicts of Interest
The authors declare that they have no conflicts of interest.
Acknowledgments
This work was supported by the Young Core Instructor Foundation from the Education Commission of Henan Province (2015GGJS-021), the Program for Science & Technology Innovation Talents in Universities of Henan Province, China (17HASTIT014), and Henan University funding (0000A40374).