- Email: [email protected]
Fabrication and visible-light photocatalytic behavior of perovskite praseodymium ferrite porous nanotubes
Journal of Power Sources 285 (2015) 178e184
Contents lists available at ScienceDirect
Journal of Power Sources journal homepage: www.elsevier.com/locate/jpowsour
Fabrication and visible-light photocatalytic behavior of perovskite praseodymium ferrite porous nanotubes Chuanxiang Qin a, b, *, Zhenyu Li b, Guoqiang Chen c, Yan Zhao b, Tong Lin b, ** a
College of Chemistry, Chemical Engineering and Materials Science, Soochow University, Suzhou 215123, China Institute for Frontier Materials, Deakin University, Geelong, VIC 3216, Australia c National Engineering Laboratory for Modern Silk, College of Textile and Clothing Engineering, Soochow University, Suzhou 215123, China b
h i g h l i g h t s Perovskite PrFeO3 porous nanotube was prepared by electrospinning and calcination. Annealing the as-spun nanofibers at a low temperature played a key role. The porous PrFeO3 tubes showed high optical absorption in the UVevisible region. The porous PrFeO3 tubes showed good visible-light photo-catalytic ability.
a r t i c l e i n f o
a b s t r a c t
Article history: Received 28 November 2014 Received in revised form 13 March 2015 Accepted 14 March 2015 Available online 17 March 2015
Perovskite praseodymium ferrite (PrFeO3) porous nanotubes are prepared by electrospinning of the precursor solution into nanofibers, subsequently by annealing the precursor fibers at a low temperature (e.g. 40 C) and finally by calcination at a high temperature. The low temperature annealing treatment is found to play a key role in the formation of porous nanotube. The porous tubes show a perovskite-type PrFeO3 crystal characteristic with high optical absorption in the UVevisible region and an energy band gap of 1.97 eV. When compared with PrFeO3 porous nanofibers and PrFeO3 particles, the PrFeO3 porous nanotubes show better visible-light photo-catalytic ability to degrade Rhodamine B in aqueous phase because of the increased surface area and more active catalytic sites on the inner walls and outer surfaces. © 2015 Elsevier B.V. All rights reserved.
Keywords: Perovskite Orthoferrites Nanotube Visible-light photocatalyst
1. Introduction Considerable interest has been devoted to developing efficient photocatalysts to deal with environment pollution over recent decades. Photocatalysts working in visible light are especially desired for the reasons that (1) visible light takes 40% of sunlight, (2) it is safer than UV light which is often used for photocatalysis degradation of pollutant, and (3) sufficient visible light is available widely from not only sunlight but also indoor lighting devices. Traditional metal oxide semiconductors (e.g. TiO2 [1], ZnO [2], SnO2 [3]) after modification show photocatalytic activity in visible light.
* Corresponding author. College of Chemistry, Chemical Engineering and Materials Science, Soochow University, Suzhou 215123, China. ** Corresponding author. Institute for Frontier Materials, Deakin University, Geelong, VIC 3216, Australia. E-mail addresses: [email protected] (C. Qin), [email protected] (T. Lin). http://dx.doi.org/10.1016/j.jpowsour.2015.03.096 0378-7753/© 2015 Elsevier B.V. All rights reserved.
However, stronger visible light photocatalytic ability was reported on some novel semiconductors including Ag3PO4 [4], Bi2WO6 [5], BiVO4 [6] and LaFeO3 [7]. Among them, the rare-earth orthoferrites (formulated as ReFeO3) with a perovskite-type crystal structure are highly promising because they are stable and environmental friendly [7e16]. LaFeO3 typically has an energy band gap in the range of 2.06e2.36 eV, regardless of the crystal structure and morphologies, facilitating visible-light photocatalysis [7e10]. In theory, photocatalysts on nano-scale have higher photocatalytic activity than the bulk counterparts because of the larger surface area and more reactive sites on the surface. Reducing catalyst size is expected to increase photocatalytic activity. However, the size decrease does not always lead to a positive result, which is in particular the case for particulate catalysts. Fibrous nanomaterials show advantages in maintaining a highly porous structure no matter whether they disperse in a liquid phase or stay in a solid state. Dong et al. [17] reported that Perovskite-type LaCoO3 nanofibers prepared by an electrospinning process and
C. Qin et al. / Journal of Power Sources 285 (2015) 178e184
subsequent thermal treatment have better photocatalytic activity than LaCoO3 particles. It is expected that photocatalytic fibers having a porous or hollow structure would have an increased surface, providing more active sites [18e24], hence high photocatalytic activities. However, to the best of our knowledge, the preparation on porous orthoferrites (ReFeO3) nanotubes and their visible-light photocatalytic performance have not been reported in research literature. Herein, we for the first time report on the preparation of a perovskite PrFeO3 porous nanotube and its visible photocatalysis property. The nanotubes were synthesized by electrospinning of a polymer solution containing inorganic oxide precursor, subsequently annealing at a low temperature and finally calcination at a high temperature. Normally, without low temperature annealing, porous fibers were prepared from this process. However, our study indicated that annealing the as-spun nanofibers at a low temperature (e.g. 40 C) played a key role in the formation porous tubes. The nanotube formation mechanism was proposed. Rhodamine B (RhB) was employed to test the photodegradation ability of PrFeO3 nanotubes in aqueous phase, which showed higher photocatalytic activity than their porous fiber and nanoparticle counterparts.
pass energy was 40 eV. UVevis diffuse absorption spectra were recorded with a Hitachi UV-3010 spectrophotometer using BaSO4 as a reference, while absorption spectra of the dye solution were obtained using a Varian Cary 3E UVevis spectrophotometer at room temperature. FTIR spectra were recorded on a Nicolet 5200 FTIR 5DX instrument. Nitrogen adsorption desorption isotherms were obtained using a Tristar 3000 apparatus. Specific surface area was calculated by the BrunauereEmmetteTeller (BET) method and the pore size distribution was calculated by the BarreteJoynereHalenda (BJH) method. 2.4. Photocatalytic activity test Photocatalytic activity was evaluated by the degradation of RhB aqueous solution under the irradiation of a 60 W desk lamp with a filter to cut off light beam of wavelength below 400 nm. In a typical test, 30 mg PrFeO3 sample and 60 mL RhB solution (5 105 mol/L) were mixed and stirred for 30 min in the dark followed by light irradiation in a hood (temperature ~25 C). The RhB concentration was measured at different irradiation time points based on the optical absorption. The photodegradation efficiency (h) was calculated according to the following equation.
2. Experimental 2.1. Materials Pr6O11 (99.9%) was purchased from Shanghai Yuelong Rare Earth New Materials, Fe (NO3)3.9H2O (98.0%), citric acid (99.5%) and poly (vinyl pyrrolidone) (PVP, Mw 1,300,000) were purchased from Sigma Aldrich. All the chemicals were used as received. The precursor solution for electrospinning was prepared as follow: 0.25 mmol Pr6O11 was dissolved in 10 mL concentrated nitric acid. After drying, greenish Pr(NO3)3.xH2O resulted. Pr(NO3)3.xH2O (1.5 mmol) was dissolved in a solution containing 8 mL DMF, 12 mL ethanol, 1.5 mmol Fe(NO3)3.9H2O and 3 mmol citric acid. 3.0 g PVP powder was then added to that metal nitrate/citric acid solution and mixed to form a homogeneous solution. The citric acid here was used as a complex agent for metal cations in the solution and helped to prevent the growth of the perovskite oxide into bulky crystals [25]. 2.2. Preparation of PrFeO3 nanotube Electrospinning was performed on a purpose-built electrospinning setup [26]. The applied voltage, spinning distance and solution flow rate were controlled at 18 kV, 22 cm and 1.6 mL/h, respectively. The as-electrospun fiber mat was placed in a blast oven at 40 C for 15 h to completely remove solvent from the fibers (the process was also referred as annealing). The sample was then calcined at different temperatures in air to produce PrFeO3. 2.3. Characterization Nanotube morphology was observed under a scanning electron microscope (SEM) using a Zeiss Supra 55 VP instrument and transmission electron microscope (TEM) on a JEOL JEM-2100. XRD patterns were recorded on a Panalytical X'PertPro X-ray diffractometer with Cu Ka radiation (l ¼ 1.54056 Å) at 40 kV and 30 mA. Thermogravimetric analysis (TGA) and differential scanning calorimeter (DSC) were measured on a Netzsch STA 409 PC/PG thermal analyzer at a heating rate of 10 C/min from room temperature to 800 C in compressed air gas flow. The X-ray photoelectron spectroscopy (XPS) spectra were obtained on a Axis Ultra HAS (Kratos) monochromatic Al Ka radiation at a reduced power of 100 W, where the step size for the high-resolution scan was 0.1 eV and the
179
h¼
C0 C 100% C0
where C0 and C are the initial RhB concentration and the residual RhB concentration after photodegradation for certain period of time. 3. Results and discussion 3.1. Preparation and morphology Scheme 1a schematically illustrates the procedure for preparing PrFeO3 nanotubes. Precursor nanofibers were produced from a homogeneous solution consisting of PVP, citric acid and metal nitrate using an electrospinning technique. PVP in the electrospinning solution functions to improve the electrospinning ability. The precursor fiber mats were subsequently subjected to an annealing treatment at 40 C and a calcination treatment at higher temperature (550e700 C). Scheme 1b shows the morphology of as-spun precursor nanofibers which look uniform with a smooth surface. As-spun fibers had a diameter in the range of 200e260 nm and a length up to several tens of micrometers (see Supporting Information Fig. S1a). The fiber mat after annealing at a blast oven at 40 C for 15 h showed no obvious change in the SEM image (see Scheme 1c). After calcination treatment at 550 C for 6 h (also marked as 550 C/6 h in this paper), sample became a yellow solid (see photo in Scheme 1a). Under SEM, fibrous structure can still be seen after the calcination treatment. However the fibers broke into short sections with a rough porous surface (Scheme 1d). The fiber diameter was reduced to 70e80 nm. The cross-section of the fibers clearly indicated that the fibers had a hollow structure (tubes). TEM imaging verified the hollow tube structure. As shown in Scheme 1e, PrFeO3 nanotube (diameter ~70 nm) has a porous inner structure with the wall thickness approximately 10 nm. The calcination condition was found to affect nanotube morphology and diameter. The tube diameter decreased from 80 nm to 50 nm with increasing the calcination temperature at the last stage. However the particle size and wall thickness inclined to increase (see Supporting Information Fig. S2). When the temperature was 700 C, beaded fiber resulted. This was attributed to the formation of larger PrFeO3 crystal particles (size ~40 nm) at such a
180
C. Qin et al. / Journal of Power Sources 285 (2015) 178e184
Scheme 1. (a) Illustration of the procedure for preparing PrFeO3 nanotubes, (bed) SEM images of nanofibers before annealing (b), after annealing treatment (c) and after calcinations (d), (e) TEM image of calcined nanotubes (inset: the SAED pattern of the corresponding sample).
high temperature. Lower temperature (e.g. 550 C) led to smaller PrFeO3 crystal particles (size ~10 nm). It was noted that the annealing process before high temperature calcination played a key role in the formation of porous tubes. Without annealing, when the as-spun fibers were subjected to high temperature calcination directly, porous fibers were formed instead of hollow nanotubes (see Supporting Information Fig. S1b). The porous fibers had a diameter around 100 nm. 3.2. DSC and TGA Differential scanning calorimeter (DSC) and thermogravimetric analysis (TGA) were used to understand the effect of the annealing and calcinations treatments on the precursor fibers. For the precursor fibers without annealing treatment, two endothermic peaks appeared at 93 C and 179 C in the DSC curve (see Fig. 1a), which corresponded to the evaporation of moisture/ethanol and DMF from the nanofibers, respectively. These peaks also appeared in the pure PVP nanofibers (non-annealed). However, for the annealed fibers, the peak for DMF disappeared, confirming the removal of the volatile solvent from the fibers. At higher temperature the annealed fiber sample showed two separate exothermic peaks at 313 C and 440 C. In comparison, the precursor fibers without annealing treatment showed a completely different peak profile to the annealed one. In the same temperature range, a broad exothermic peak was observed instead of two distinct peaks. For the pure PVP fiber sample, showed one obvious exothermic peak at 315 C. TGA curves provide information about weight loss during heat treatment. As shown in Fig. 1b, pure PVP nanofibers without any additive shows three main weight loss steps (also see details in the Supporting Information Fig. S3). The first one was at around 75 C, which corresponded to desorption of water & ethanol molecules from the fibers. The second weight loss happened at around 310 C, corresponding to the degradation of PVP [27], and the third one started at around 493 C, related to the further degradation of PVP residual. For the annealed fibers, less weight loss happened in the temperature 140e160 C, and weight loss in the higher temperature showed two distinct steps. The one in the temperature 270e350 C was associated with the decomposition of citric acid and PVP, and another at around 429 C was assigned to the formation of PrFeO3 from metal nitrate. However, for the precursor
Fig. 1. (a) DSC and (b) TGA curves of different fiber samples in air (10 C/min). (i) Pure PVP nanofibers, (ii) Precursor nanofibers after annealing at 40 C for 15 h, (iii) Precursor nanofibers without annealing treatment.
C. Qin et al. / Journal of Power Sources 285 (2015) 178e184
181
fibers without annealing, one cannot find the obvious characteristic weight loss of PVP material. The noticeable difference in the weight loss in the higher temperature range confirmed the significant effect of low temperature annealing on the precursor nanofibers. 3.3. Nanotube formation mechanism Based on above results, the formation mechanism of porous nanotubes was proposed, as illustration in Scheme 1a. When the as-spun nanofibers were annealed at 40 C, the evaporation of the residual solvent and moisture from the fibers induced migration of smaller molecules, such as nitrate and citric acid, to the surface layer of the fibers because of their high mobility. This led to uneven distribution of chemical components between the sheath layer and the core of the fiber. The metal nitrate and citric acid enriched in the surface layer, leaving much lower concentration in the core area. After removal of PVP through calcination, structural collapse took place in the core, while porous sheath layer was formed and the thermal decomposition of metal nitrates and citric acid resulted in porous PrFeO3 sheath [6]. 3.4. Nitrogen adsorption Table 1 shows specific surface area and pore size based on the nitrogen adsorptionedesorption isotherms and the corresponding adsorption spectra are provided in the Supporting Information (Figs. S5, S6). The hysteresis loop in the high-pressure range (0.7 < P/P0 < 1) can be associated with the larger pores formed between secondary particles due to aggregation [28]. For comparison, porous PrFeO3 fibers prepared from the precursor nanofibers without annealing treatment and PrFeO3 particles prepared under the same conditions (see preparation details in the Supporting Information S1) were also subjected to nitrogen adsorptiondesorption test. Among the samples tested, PrFeO3 nanotubes had the largest specific surface area (16.9 m2;/g), attributing to their hollow, porous structure. The calcination condition showed an influence on the surface area and pore volume. The specific surface area and pore volume decreased with increasing the calcination temperature. 3.5. Chemical components XPS and FTIR spectra were used to examine the chemical composition of PrFeO3 samples. Fig. 2a shows the XPS survey spectrum of PrFeO3 nanotubes. The carbon peak was attributed to adventitious carbon on the surface of the sample [28]. The O 1s XPS spectrum corresponds to two kinds of O chemical states, including crystal lattice oxygen (OL, attributed to the contribution of PreO and FeeO in PrFeO3 crystal lattice) and hydroxyl oxygen (OH, related to the hydroxyl groups resulting mainly from the chemisorbed water) with increasing binding energy [9]. The binding
Table 1 Specific surface area, pore volume and photodegradation efficiency of PrFeO3. Calcination condition Porous nanotubes
Porous nanofibers Particles
550 550 600 700 550 550
C/4 C/6 C/4 C/4 C/6 C/6
h h h h h h
Surface area (m2/g)
Pore volume (cm3/g)
ha (%)
20.6 16.9 14.9 10.4 15.5 6.4
0.059 0.048 0.035 0.019 0.033 0.009
84.3 90.6 73.3 45.7 75.4 41.4
a In each test, an aqueous RhB solution (60 mL, 5 105 mol/L) containing PrFeO3 sample (30 mg) was irradiated by visible light for 30 h.
Fig. 2. XPS spectra of porous PrFeO3 nanotubes (550 C/6 h).
energy of Fe 2p3/2 and Fe 2p1/2 was observed at 707.1 eV and 720.9 eV (Fig. 2b), indicating the oxide form of the Fe3þ [9,28]. Fig. 2c shows two strong peaks at 929.8 eV and 950.4 eV, which are assigned to spineorbit splitting of 3d5/2 and 3d3/2 of Pr3þ ions in oxide form. The binding energy of Fe 2p and Pr 3d is in good agreement with the literature report [29]. The FTIR spectra (see Supporting Information Fig. S7) show that pure rare-earth orthoferrites have no obvious vibration bands in
182
C. Qin et al. / Journal of Power Sources 285 (2015) 178e184
the region 4000e800 cm1. Strong vibration bands were observed around 548 cm1, corresponding to FeeO stretching vibration [10,30].
Table 2 Refined crystallographic parameters of PrFeO3.
3.6. Crystal structure Fig. 3 shows the XRD patterns of the electrospun nanofibers before and after calcination treatment. Without calcination, the asspun fibers only showed a broad band at 2q ¼ 21, which was ascribed to the semi-crystalline PVP. After calcination, such a broad peak disappeared, and well defined diffraction peaks appeared. The patterns aligned well with the PDF2 standard card No. 78-2424 (perovskite-type PrFeO3) selected in the International Centre for Diffraction Data (ICDD) database. No impurity peaks were observed in the XRD spectra. For the sample calcined at 550 C for 4 h, the diffraction peaks began to match the PDF standard card, though the wider peaks and lower intensity. When the dwelling time in 550 C increased to 6 h, all the main peaks (22.7(110), 32.4(112), 40(202), 46(220), 58.1(204)) matched well with the PDF standard card (Fig. 3a). The
Formula
PrFeO3
Radiation 2q range (degree) Symmetry Space group# a/Å b/Å c/Å a/ b/ g/ Z Rp Rwp X2 V/Å3
Cu/Ka 10e80 Orthorhombic Pbnm (62) 5.4826 (12) 5.5714 (12) 7.7871 (19) 90 90 90 4 0.0857 0.1112 1.301 237.86 (9)
diffraction rings are discontinuous as shown in the insert image of TEM image (Scheme 1e), indicating good crystallinity of PrFeO3 nanoparticles as observed from the selection area diffraction pattern (SAED). The major diffraction spots correspond to (110), (112), (202), (220) and (204) diffraction patterns of PrFeO3 with perovskite structure [7]. With the further increase of calcination temperature, the peak intensity increased, indicating that increasing temperature facilitates the crystallization of PrFeO3. Based on the experimental XRD pattern, the structure parameters and atom positions of PrFeO3 in the nanotubes were refined by the GSAS program, as shown in Fig. 3b. The sample is in an orthorhombic system with the space group of Pbnm (62) (unit parameters a ¼ 5.4826(12) Å, b ¼ 5.5714(12) Å, c ¼ 7.7871(5) Å, a ¼ b ¼ g ¼ 90 and V ¼ 237.86(9) Å3), as shown in Table 2. No impurity lines were observed, and this was in agreement with Pandey et al.'s report [29]. The inset shows the sketch map of the PrFeO3 unit cells, which was modeled using the Diamond Crystal and Molecular Structure Visualization software based on the atomic coordinate's refinement data in Table 3. The structure is derived from the typical perovskite (ABO3), where Pr3þ (A-site) ions are coordinated to 12 oxygen atoms and Fe3þ ions (coordinated to 6 oxygen atoms) are ordered in the B-site. The unit cell contains four equivalent Fe3þ ions situated at octahedral centers formed by six nearest-neighbor oxygen atoms, while the tilting of the octahedral axes with respect to the c-axis is minimal [31]. 3.7. Band gap The energy band is a key feature reflecting the photocatalytic activity of a photocatalyst material [7]. UVevis absorption spectrum of PrFeO3 nanotubes (550 C/6 h) was tested (see Supporting Information Fig. S8). One can see that PrFeO3 nanotubes exhibit obvious absorption in visible light regions, which can mainly be attributed to the band gap electronic transition from the valence band to the conduction band (O 2p/Fe 3d) [7]. Based on the data of absorption spectra, the band gap can be calculated by using the formula (band gap ðeVÞ ¼ ð1240=lg Þ nm [15]; where lg is the absorption wavelength value obtained from absorbance spectrum).
Table 3 Refined atomic coordinate parameters of PrFeO3.
Fig. 3. (a) XRD patterns of electrospun nanofibers before and after calcination treatment, (b) representative experimental (crossed) and calculated (red solid line) X-ray diffraction profiles of PrFeO3 nanotubes (inset is the schematic views of the structure geometries along the c-direction). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Atom
x/a
y/b
z/c
U[Å2]
Pr1 Fe1 O1 O2
0.9946 (18) 0.00000 0.094 (9) 0.716 (6)
0.0408 (5) 0.50000 0.475 (5) 0.301 (5)
0.25000 0.00000 0.25000 0.040 (5)
0.0045 0.0078 0.029 0.022
(11) (18) (13) (10)
C. Qin et al. / Journal of Power Sources 285 (2015) 178e184
183
The band gap of the porous PrFeO3 nanotubes was 1.97 eV (lg ¼ 629 nm). This value is similar to that of nanocrystalline PrFeO3 synthesized by a combustion method (1.88 eV), template method (2.03 eV) and solegel method (2.08 eV) [15]. The absorption spans to the wavelength of 629 nm also indicates that PrFeO3 nanotubes could be used for visible-light photocatalysis [7]. 3.8. Photocatalytic activity To prove the photocatalytic activity in visible light, PrFeO3 nanotubes were placed in an aqueous solution containing RhB, and a desk lamp (60 W) with a UV filter was used as a light source to irradiate the solution. Fig. 4a shows the UVevis spectra taken during the photodegradation test. The peaks at 553 nm and 503 nm are the characteristic absorption of tetraethylated rhodamine and the absorption of fully de-ethylated RhB molecule [5]. Irradiation by the desk lamp, the solution color changed gradually from red to light green-yellow and finally colorless. The inset images show the color change of the RhB solution. After 30 h of irradiation, 90.6% of RhB was decolorized and the solution turned almost colorless, indicating that chromophoric structure was destroyed [5]. It was reported that photocatalysts in fibrous form were easier to separate from reaction solution than nanoparticles [17,32]. Here, PrFeO3 nanotubes were recycled by centrifuge for 5 min and its repeating usability was tested by decolorization of fresh RhB solutions with the used catalyst. As shown in Fig. 4b, in the 1st, the 2nd and the 3rd test cycles, the decolorization efficiency of 90.6%, 87.9% and 86.3% was achieved respectively. The results indicate that the PrFeO3 nanotubes can be used for multicycle photodegradation. The Supporting Information in Fig. S9 displays the XRD patterns of the PrFeO3 nanotubes before and after the three cycles of photodegradation. No obvious change could be found between the two curves, which indicated that the PrFeO3 nanotubes were not photo corroded during the photodegradation reaction. Fig. 4c shows the photo-degradation efficiency of RhB using different PrFeO3. 46.5% RhB is degraded by the nanotube sample in 6 h, while for the porous fiber and nanoparticle samples, the decolorization efficiency was 29.5% and 15.7%, respectively. And all the residual RhB aqueous were still in red. 12 h later, the degradation efficiency of 56.2%, 49.4% and 24.2% was achieved for the porous nanotube, porous nanofiber and nanoparticle samples, respectively. The residual RhB solutions were green-yellow, light red and red. 90.6% RhB was degraded by the nanotube sample in 30 h, while for the porous fiber and nanoparticle samples, 24.6% and 58.6% of RhB were still left, respectively. The nanotube sample exhibited the highest efficiency for its highest specific surface area [20] (see Table 1). The photodegradation efficiency of PrFeO3 nanotubes calcinated at different temperature was also tested (see Supporting Information Fig. S10). The efficiency decreased when the temperature increased from 550 to 700 C and the highest one was the sample calcined at 550 C for 6 h. This can be explained by that the sample (550 C/6 h) has the suitable crystallinity and specific surface area which are two important parameters determining photocatalytic activity. The sample (550 C/4 h) exhibited slightly lower efficiency than the sample (550 C/6 h) for poor crystallinity (see XRD spectrum in Fig. 3a) although it has the highest specific surface area. 4. Conclusions
Fig. 4. (a) UVevis spectral change of aqueous RhB solution at different photodegradation (PrFeO3 nanotubes, 550 C/6 h), (b) repeatability tests on the sample, and (c) photodegradation efficiency (h) of RhB solution in presence and absence of different PrFeO3 photocatalysts.
Perovskite-type PrFeO3 porous nanotubes with diameter 70e80 nm and wall thickness around 10 nm have been prepared successfully. The low temperature annealing treatment of precursor nanofibers was found to play a key role in determining the formation of porous PrFeO3 nanotubes. The porous tubes showed a
perovskite-type PrFeO3 crystal characteristic with a high absorption in the UVevisible light region and an energy band gap as low as 1.97 eV. When compared with porous PrFeO3 nanofibers and PrFeO3 particles, the PrFeO3 porous nanotubes showed higher
184
C. Qin et al. / Journal of Power Sources 285 (2015) 178e184
visible-light photocatalytic ability to degrade Rhodamine B in aqueous phase because of the increased specific surface areas and more active catalytic sites on the inner walls and outer surfaces. Acknowledgments The authors thank financial support from National Natural Science Foundation of China (No. 51273134), Jiangsu Provincial Natural Science Foundation of China (No. BK2012635) and Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD). Alfred Deakin Postdoctoral Fellowship awarded to the 1st author is also acknowledged. Appendix A. Supplementary data Supplementary data related to this article can be found online at http://dx.doi.org/10.1016/j.jpowsour.2015.03.096. References [1] S. Li, Y. Guo, L. Zhang, J. Wang, Y. Li, Y. Li, B. Wang, J. Power Sources 252 (2014) 21e27. [2] M. Basu, N. Garg, A.K. Ganguli, J. Mater. Chem. A 2 (2014) 7517e7525. [3] S.A. Ansari, M.M. Khan, M.O. Ansari, J. Lee, M.H. Cho, RSC Adv. 4 (2014) 26013e26021. [4] Y. Bi, S. Ouyang, J. Cao, J. Ye, Phys. Chem. Chem. Phys. 13 (2011) 10071e10075. [5] H. Fu, C. Pan, W. Yao, Y. Zhu, J. Phys. Chem. B 109 (2005) 22432e22439. [6] Y. Cheng, J. Chen, X. Yan, Z. Zheng, Q. Xue, RSC Adv. 3 (2013) 20606e20612. [7] S. Thirumalairajan, K. Girija, I. Ganesh, D. Mangalaraj, C. Viswanathan, A. Balamurugan, N. Ponpandian, Chem. Eng. J. 209 (2012) 420e428. [8] P. Tang, Y. Tong, H. Chen, F. Cao, G. Pan, Curr. Appl. Phys. 13 (2013) 340e343.
[9] K.M. Parida, K.H. Reddy, S. Martha, D.P. Das, N. Biswal, Int. J. Hydrogen Energy 35 (2010) 12161e12168. [10] S.N. Tijare, M.V. Joshi, P.S. Padole, P.A. Mangrulkar, S.S. Rayalu, N.K. Labhsetwar, Int. J. Hydrogen Energy 37 (2012) 10451e10456. [11] J. Ding, X. Lü, H. Shu, J. Xie, H. Zhang, Mater. Sci. Eng. B 171 (2010) 31e34. [12] L. Ju, Z. Chen, L. Fang, W. Dong, F. Zheng, M. Shen, J. Am. Ceram. Soc. 94 (2011) 3418e3424. [13] H. Yang, J.X. Zhang, G.J. Lin, T. Xian, J.L. Jiang, Adv. Powder Technol. 24 (2013) 242e245. [14] L. Li, X. Wang, Y. Lan, W. Gu, S. Zhang, Ind. Eng. Chem. Res. 52 (2013) 9130e9136. [15] S.N. Tijare, S. Bakardjieva, J. Subrt, M.V. Joshi, S.S. Rayalu, S. Hishita, N. Labhsetwar, J. Chem. Sci. 126 (2014) 517e525. [16] G-J. Lin, H. Yang, T. Xian, Chem. J. Chin. U 33 (2012) 1565e1571. [17] B. Dong, Z. Li, Z. Li, X. Xu, M. Song, W. Zheng, C. Wang, S.S. Al-Deyab, M. ElNewehy, J. Am. Ceram. Soc. 93 (2010) 3587e3590. [18] D. Hou, X. Hu, Y. Wen, B. Shan, P. Hu, X. Xiong, Y. Qiao, Y. Huang, Phys. Chem. Chem. Phys. 15 (2013) 20698e20705. [19] X. Li, C. Tang, M. Ai, L. Dong, Z. Xu, Chem. Mater. 22 (2010) 4879e4889. [20] S. Mohan, B. Subramanian, RSC Adv. 3 (2013) 23737e23744. [21] B. Lu, C. Zhu, Z. Zhang, W. Lan, E. Xie, J. Mater. Chem. 22 (2012) 1375e1379. [22] Y. Cheng, W. Huang, Y. Zhang, L. Zhu, Y. Liu, X. Fan, X. Cao, CrystEngComm 12 (2010) 2256e2260. [23] S. Zhan, D. Chen, X. Jiao, C. Tao, J. Phys. Chem. B 110 (2006) 11199e11204. [24] Z. Dong, S.J. Kennedy, Y. Wu, J. Power Sources 196 (2011) 4886e4904. [25] C.-Q. Chen, W. Li, C.-Y. Cao, W.-G. Song, J. Mater. Chem. 20 (2010) 6968e6974. [26] Z. Li, X. Wang, T. Lin, J. Mater. Chem. A 2 (2014) 13655e13660. [27] J.P. Cheng, B.B. Wang, M.G. Zhao, F. Liu, X.B. Zhang, Sens. Actuat. B: Chem. 190 (2014) 78e85. [28] S. Thirumalairajan, K. Girija, N.Y. Hebalkar, D. Mangalaraj, C. Viswanathan, N. Ponpandian, RSC Adv. 3 (2013) 7549e7561. [29] S.K. Pandey, R. Bindu, P. Bhatt, S.M. Chaudhari, A.V. Pimpale, Phys. B: Condens. Matter 365 (2005) 47e54. [30] M. Popa, J.M.C. Moreno, J. Alloy Compd. 509 (2011) 4108e4116. ndez-García, G. Colo n, Chem. Rev. 112 (2011) [31] A. Kubacka, M. Ferna 1555e1614. [32] D. Lin, H. Wu, R. Zhang, W. Pan, Chem. Mater. 21 (2009) 3479e3484.
Contents lists available at ScienceDirect
Journal of Power Sources journal homepage: www.elsevier.com/locate/jpowsour
Fabrication and visible-light photocatalytic behavior of perovskite praseodymium ferrite porous nanotubes Chuanxiang Qin a, b, *, Zhenyu Li b, Guoqiang Chen c, Yan Zhao b, Tong Lin b, ** a
College of Chemistry, Chemical Engineering and Materials Science, Soochow University, Suzhou 215123, China Institute for Frontier Materials, Deakin University, Geelong, VIC 3216, Australia c National Engineering Laboratory for Modern Silk, College of Textile and Clothing Engineering, Soochow University, Suzhou 215123, China b
h i g h l i g h t s Perovskite PrFeO3 porous nanotube was prepared by electrospinning and calcination. Annealing the as-spun nanofibers at a low temperature played a key role. The porous PrFeO3 tubes showed high optical absorption in the UVevisible region. The porous PrFeO3 tubes showed good visible-light photo-catalytic ability.
a r t i c l e i n f o
a b s t r a c t
Article history: Received 28 November 2014 Received in revised form 13 March 2015 Accepted 14 March 2015 Available online 17 March 2015
Perovskite praseodymium ferrite (PrFeO3) porous nanotubes are prepared by electrospinning of the precursor solution into nanofibers, subsequently by annealing the precursor fibers at a low temperature (e.g. 40 C) and finally by calcination at a high temperature. The low temperature annealing treatment is found to play a key role in the formation of porous nanotube. The porous tubes show a perovskite-type PrFeO3 crystal characteristic with high optical absorption in the UVevisible region and an energy band gap of 1.97 eV. When compared with PrFeO3 porous nanofibers and PrFeO3 particles, the PrFeO3 porous nanotubes show better visible-light photo-catalytic ability to degrade Rhodamine B in aqueous phase because of the increased surface area and more active catalytic sites on the inner walls and outer surfaces. © 2015 Elsevier B.V. All rights reserved.
Keywords: Perovskite Orthoferrites Nanotube Visible-light photocatalyst
1. Introduction Considerable interest has been devoted to developing efficient photocatalysts to deal with environment pollution over recent decades. Photocatalysts working in visible light are especially desired for the reasons that (1) visible light takes 40% of sunlight, (2) it is safer than UV light which is often used for photocatalysis degradation of pollutant, and (3) sufficient visible light is available widely from not only sunlight but also indoor lighting devices. Traditional metal oxide semiconductors (e.g. TiO2 [1], ZnO [2], SnO2 [3]) after modification show photocatalytic activity in visible light.
* Corresponding author. College of Chemistry, Chemical Engineering and Materials Science, Soochow University, Suzhou 215123, China. ** Corresponding author. Institute for Frontier Materials, Deakin University, Geelong, VIC 3216, Australia. E-mail addresses: [email protected] (C. Qin), [email protected] (T. Lin). http://dx.doi.org/10.1016/j.jpowsour.2015.03.096 0378-7753/© 2015 Elsevier B.V. All rights reserved.
However, stronger visible light photocatalytic ability was reported on some novel semiconductors including Ag3PO4 [4], Bi2WO6 [5], BiVO4 [6] and LaFeO3 [7]. Among them, the rare-earth orthoferrites (formulated as ReFeO3) with a perovskite-type crystal structure are highly promising because they are stable and environmental friendly [7e16]. LaFeO3 typically has an energy band gap in the range of 2.06e2.36 eV, regardless of the crystal structure and morphologies, facilitating visible-light photocatalysis [7e10]. In theory, photocatalysts on nano-scale have higher photocatalytic activity than the bulk counterparts because of the larger surface area and more reactive sites on the surface. Reducing catalyst size is expected to increase photocatalytic activity. However, the size decrease does not always lead to a positive result, which is in particular the case for particulate catalysts. Fibrous nanomaterials show advantages in maintaining a highly porous structure no matter whether they disperse in a liquid phase or stay in a solid state. Dong et al. [17] reported that Perovskite-type LaCoO3 nanofibers prepared by an electrospinning process and
C. Qin et al. / Journal of Power Sources 285 (2015) 178e184
subsequent thermal treatment have better photocatalytic activity than LaCoO3 particles. It is expected that photocatalytic fibers having a porous or hollow structure would have an increased surface, providing more active sites [18e24], hence high photocatalytic activities. However, to the best of our knowledge, the preparation on porous orthoferrites (ReFeO3) nanotubes and their visible-light photocatalytic performance have not been reported in research literature. Herein, we for the first time report on the preparation of a perovskite PrFeO3 porous nanotube and its visible photocatalysis property. The nanotubes were synthesized by electrospinning of a polymer solution containing inorganic oxide precursor, subsequently annealing at a low temperature and finally calcination at a high temperature. Normally, without low temperature annealing, porous fibers were prepared from this process. However, our study indicated that annealing the as-spun nanofibers at a low temperature (e.g. 40 C) played a key role in the formation porous tubes. The nanotube formation mechanism was proposed. Rhodamine B (RhB) was employed to test the photodegradation ability of PrFeO3 nanotubes in aqueous phase, which showed higher photocatalytic activity than their porous fiber and nanoparticle counterparts.
pass energy was 40 eV. UVevis diffuse absorption spectra were recorded with a Hitachi UV-3010 spectrophotometer using BaSO4 as a reference, while absorption spectra of the dye solution were obtained using a Varian Cary 3E UVevis spectrophotometer at room temperature. FTIR spectra were recorded on a Nicolet 5200 FTIR 5DX instrument. Nitrogen adsorption desorption isotherms were obtained using a Tristar 3000 apparatus. Specific surface area was calculated by the BrunauereEmmetteTeller (BET) method and the pore size distribution was calculated by the BarreteJoynereHalenda (BJH) method. 2.4. Photocatalytic activity test Photocatalytic activity was evaluated by the degradation of RhB aqueous solution under the irradiation of a 60 W desk lamp with a filter to cut off light beam of wavelength below 400 nm. In a typical test, 30 mg PrFeO3 sample and 60 mL RhB solution (5 105 mol/L) were mixed and stirred for 30 min in the dark followed by light irradiation in a hood (temperature ~25 C). The RhB concentration was measured at different irradiation time points based on the optical absorption. The photodegradation efficiency (h) was calculated according to the following equation.
2. Experimental 2.1. Materials Pr6O11 (99.9%) was purchased from Shanghai Yuelong Rare Earth New Materials, Fe (NO3)3.9H2O (98.0%), citric acid (99.5%) and poly (vinyl pyrrolidone) (PVP, Mw 1,300,000) were purchased from Sigma Aldrich. All the chemicals were used as received. The precursor solution for electrospinning was prepared as follow: 0.25 mmol Pr6O11 was dissolved in 10 mL concentrated nitric acid. After drying, greenish Pr(NO3)3.xH2O resulted. Pr(NO3)3.xH2O (1.5 mmol) was dissolved in a solution containing 8 mL DMF, 12 mL ethanol, 1.5 mmol Fe(NO3)3.9H2O and 3 mmol citric acid. 3.0 g PVP powder was then added to that metal nitrate/citric acid solution and mixed to form a homogeneous solution. The citric acid here was used as a complex agent for metal cations in the solution and helped to prevent the growth of the perovskite oxide into bulky crystals [25]. 2.2. Preparation of PrFeO3 nanotube Electrospinning was performed on a purpose-built electrospinning setup [26]. The applied voltage, spinning distance and solution flow rate were controlled at 18 kV, 22 cm and 1.6 mL/h, respectively. The as-electrospun fiber mat was placed in a blast oven at 40 C for 15 h to completely remove solvent from the fibers (the process was also referred as annealing). The sample was then calcined at different temperatures in air to produce PrFeO3. 2.3. Characterization Nanotube morphology was observed under a scanning electron microscope (SEM) using a Zeiss Supra 55 VP instrument and transmission electron microscope (TEM) on a JEOL JEM-2100. XRD patterns were recorded on a Panalytical X'PertPro X-ray diffractometer with Cu Ka radiation (l ¼ 1.54056 Å) at 40 kV and 30 mA. Thermogravimetric analysis (TGA) and differential scanning calorimeter (DSC) were measured on a Netzsch STA 409 PC/PG thermal analyzer at a heating rate of 10 C/min from room temperature to 800 C in compressed air gas flow. The X-ray photoelectron spectroscopy (XPS) spectra were obtained on a Axis Ultra HAS (Kratos) monochromatic Al Ka radiation at a reduced power of 100 W, where the step size for the high-resolution scan was 0.1 eV and the
179
h¼
C0 C 100% C0
where C0 and C are the initial RhB concentration and the residual RhB concentration after photodegradation for certain period of time. 3. Results and discussion 3.1. Preparation and morphology Scheme 1a schematically illustrates the procedure for preparing PrFeO3 nanotubes. Precursor nanofibers were produced from a homogeneous solution consisting of PVP, citric acid and metal nitrate using an electrospinning technique. PVP in the electrospinning solution functions to improve the electrospinning ability. The precursor fiber mats were subsequently subjected to an annealing treatment at 40 C and a calcination treatment at higher temperature (550e700 C). Scheme 1b shows the morphology of as-spun precursor nanofibers which look uniform with a smooth surface. As-spun fibers had a diameter in the range of 200e260 nm and a length up to several tens of micrometers (see Supporting Information Fig. S1a). The fiber mat after annealing at a blast oven at 40 C for 15 h showed no obvious change in the SEM image (see Scheme 1c). After calcination treatment at 550 C for 6 h (also marked as 550 C/6 h in this paper), sample became a yellow solid (see photo in Scheme 1a). Under SEM, fibrous structure can still be seen after the calcination treatment. However the fibers broke into short sections with a rough porous surface (Scheme 1d). The fiber diameter was reduced to 70e80 nm. The cross-section of the fibers clearly indicated that the fibers had a hollow structure (tubes). TEM imaging verified the hollow tube structure. As shown in Scheme 1e, PrFeO3 nanotube (diameter ~70 nm) has a porous inner structure with the wall thickness approximately 10 nm. The calcination condition was found to affect nanotube morphology and diameter. The tube diameter decreased from 80 nm to 50 nm with increasing the calcination temperature at the last stage. However the particle size and wall thickness inclined to increase (see Supporting Information Fig. S2). When the temperature was 700 C, beaded fiber resulted. This was attributed to the formation of larger PrFeO3 crystal particles (size ~40 nm) at such a
180
C. Qin et al. / Journal of Power Sources 285 (2015) 178e184
Scheme 1. (a) Illustration of the procedure for preparing PrFeO3 nanotubes, (bed) SEM images of nanofibers before annealing (b), after annealing treatment (c) and after calcinations (d), (e) TEM image of calcined nanotubes (inset: the SAED pattern of the corresponding sample).
high temperature. Lower temperature (e.g. 550 C) led to smaller PrFeO3 crystal particles (size ~10 nm). It was noted that the annealing process before high temperature calcination played a key role in the formation of porous tubes. Without annealing, when the as-spun fibers were subjected to high temperature calcination directly, porous fibers were formed instead of hollow nanotubes (see Supporting Information Fig. S1b). The porous fibers had a diameter around 100 nm. 3.2. DSC and TGA Differential scanning calorimeter (DSC) and thermogravimetric analysis (TGA) were used to understand the effect of the annealing and calcinations treatments on the precursor fibers. For the precursor fibers without annealing treatment, two endothermic peaks appeared at 93 C and 179 C in the DSC curve (see Fig. 1a), which corresponded to the evaporation of moisture/ethanol and DMF from the nanofibers, respectively. These peaks also appeared in the pure PVP nanofibers (non-annealed). However, for the annealed fibers, the peak for DMF disappeared, confirming the removal of the volatile solvent from the fibers. At higher temperature the annealed fiber sample showed two separate exothermic peaks at 313 C and 440 C. In comparison, the precursor fibers without annealing treatment showed a completely different peak profile to the annealed one. In the same temperature range, a broad exothermic peak was observed instead of two distinct peaks. For the pure PVP fiber sample, showed one obvious exothermic peak at 315 C. TGA curves provide information about weight loss during heat treatment. As shown in Fig. 1b, pure PVP nanofibers without any additive shows three main weight loss steps (also see details in the Supporting Information Fig. S3). The first one was at around 75 C, which corresponded to desorption of water & ethanol molecules from the fibers. The second weight loss happened at around 310 C, corresponding to the degradation of PVP [27], and the third one started at around 493 C, related to the further degradation of PVP residual. For the annealed fibers, less weight loss happened in the temperature 140e160 C, and weight loss in the higher temperature showed two distinct steps. The one in the temperature 270e350 C was associated with the decomposition of citric acid and PVP, and another at around 429 C was assigned to the formation of PrFeO3 from metal nitrate. However, for the precursor
Fig. 1. (a) DSC and (b) TGA curves of different fiber samples in air (10 C/min). (i) Pure PVP nanofibers, (ii) Precursor nanofibers after annealing at 40 C for 15 h, (iii) Precursor nanofibers without annealing treatment.
C. Qin et al. / Journal of Power Sources 285 (2015) 178e184
181
fibers without annealing, one cannot find the obvious characteristic weight loss of PVP material. The noticeable difference in the weight loss in the higher temperature range confirmed the significant effect of low temperature annealing on the precursor nanofibers. 3.3. Nanotube formation mechanism Based on above results, the formation mechanism of porous nanotubes was proposed, as illustration in Scheme 1a. When the as-spun nanofibers were annealed at 40 C, the evaporation of the residual solvent and moisture from the fibers induced migration of smaller molecules, such as nitrate and citric acid, to the surface layer of the fibers because of their high mobility. This led to uneven distribution of chemical components between the sheath layer and the core of the fiber. The metal nitrate and citric acid enriched in the surface layer, leaving much lower concentration in the core area. After removal of PVP through calcination, structural collapse took place in the core, while porous sheath layer was formed and the thermal decomposition of metal nitrates and citric acid resulted in porous PrFeO3 sheath [6]. 3.4. Nitrogen adsorption Table 1 shows specific surface area and pore size based on the nitrogen adsorptionedesorption isotherms and the corresponding adsorption spectra are provided in the Supporting Information (Figs. S5, S6). The hysteresis loop in the high-pressure range (0.7 < P/P0 < 1) can be associated with the larger pores formed between secondary particles due to aggregation [28]. For comparison, porous PrFeO3 fibers prepared from the precursor nanofibers without annealing treatment and PrFeO3 particles prepared under the same conditions (see preparation details in the Supporting Information S1) were also subjected to nitrogen adsorptiondesorption test. Among the samples tested, PrFeO3 nanotubes had the largest specific surface area (16.9 m2;/g), attributing to their hollow, porous structure. The calcination condition showed an influence on the surface area and pore volume. The specific surface area and pore volume decreased with increasing the calcination temperature. 3.5. Chemical components XPS and FTIR spectra were used to examine the chemical composition of PrFeO3 samples. Fig. 2a shows the XPS survey spectrum of PrFeO3 nanotubes. The carbon peak was attributed to adventitious carbon on the surface of the sample [28]. The O 1s XPS spectrum corresponds to two kinds of O chemical states, including crystal lattice oxygen (OL, attributed to the contribution of PreO and FeeO in PrFeO3 crystal lattice) and hydroxyl oxygen (OH, related to the hydroxyl groups resulting mainly from the chemisorbed water) with increasing binding energy [9]. The binding
Table 1 Specific surface area, pore volume and photodegradation efficiency of PrFeO3. Calcination condition Porous nanotubes
Porous nanofibers Particles
550 550 600 700 550 550
C/4 C/6 C/4 C/4 C/6 C/6
h h h h h h
Surface area (m2/g)
Pore volume (cm3/g)
ha (%)
20.6 16.9 14.9 10.4 15.5 6.4
0.059 0.048 0.035 0.019 0.033 0.009
84.3 90.6 73.3 45.7 75.4 41.4
a In each test, an aqueous RhB solution (60 mL, 5 105 mol/L) containing PrFeO3 sample (30 mg) was irradiated by visible light for 30 h.
Fig. 2. XPS spectra of porous PrFeO3 nanotubes (550 C/6 h).
energy of Fe 2p3/2 and Fe 2p1/2 was observed at 707.1 eV and 720.9 eV (Fig. 2b), indicating the oxide form of the Fe3þ [9,28]. Fig. 2c shows two strong peaks at 929.8 eV and 950.4 eV, which are assigned to spineorbit splitting of 3d5/2 and 3d3/2 of Pr3þ ions in oxide form. The binding energy of Fe 2p and Pr 3d is in good agreement with the literature report [29]. The FTIR spectra (see Supporting Information Fig. S7) show that pure rare-earth orthoferrites have no obvious vibration bands in
182
C. Qin et al. / Journal of Power Sources 285 (2015) 178e184
the region 4000e800 cm1. Strong vibration bands were observed around 548 cm1, corresponding to FeeO stretching vibration [10,30].
Table 2 Refined crystallographic parameters of PrFeO3.
3.6. Crystal structure Fig. 3 shows the XRD patterns of the electrospun nanofibers before and after calcination treatment. Without calcination, the asspun fibers only showed a broad band at 2q ¼ 21, which was ascribed to the semi-crystalline PVP. After calcination, such a broad peak disappeared, and well defined diffraction peaks appeared. The patterns aligned well with the PDF2 standard card No. 78-2424 (perovskite-type PrFeO3) selected in the International Centre for Diffraction Data (ICDD) database. No impurity peaks were observed in the XRD spectra. For the sample calcined at 550 C for 4 h, the diffraction peaks began to match the PDF standard card, though the wider peaks and lower intensity. When the dwelling time in 550 C increased to 6 h, all the main peaks (22.7(110), 32.4(112), 40(202), 46(220), 58.1(204)) matched well with the PDF standard card (Fig. 3a). The
Formula
PrFeO3
Radiation 2q range (degree) Symmetry Space group# a/Å b/Å c/Å a/ b/ g/ Z Rp Rwp X2 V/Å3
Cu/Ka 10e80 Orthorhombic Pbnm (62) 5.4826 (12) 5.5714 (12) 7.7871 (19) 90 90 90 4 0.0857 0.1112 1.301 237.86 (9)
diffraction rings are discontinuous as shown in the insert image of TEM image (Scheme 1e), indicating good crystallinity of PrFeO3 nanoparticles as observed from the selection area diffraction pattern (SAED). The major diffraction spots correspond to (110), (112), (202), (220) and (204) diffraction patterns of PrFeO3 with perovskite structure [7]. With the further increase of calcination temperature, the peak intensity increased, indicating that increasing temperature facilitates the crystallization of PrFeO3. Based on the experimental XRD pattern, the structure parameters and atom positions of PrFeO3 in the nanotubes were refined by the GSAS program, as shown in Fig. 3b. The sample is in an orthorhombic system with the space group of Pbnm (62) (unit parameters a ¼ 5.4826(12) Å, b ¼ 5.5714(12) Å, c ¼ 7.7871(5) Å, a ¼ b ¼ g ¼ 90 and V ¼ 237.86(9) Å3), as shown in Table 2. No impurity lines were observed, and this was in agreement with Pandey et al.'s report [29]. The inset shows the sketch map of the PrFeO3 unit cells, which was modeled using the Diamond Crystal and Molecular Structure Visualization software based on the atomic coordinate's refinement data in Table 3. The structure is derived from the typical perovskite (ABO3), where Pr3þ (A-site) ions are coordinated to 12 oxygen atoms and Fe3þ ions (coordinated to 6 oxygen atoms) are ordered in the B-site. The unit cell contains four equivalent Fe3þ ions situated at octahedral centers formed by six nearest-neighbor oxygen atoms, while the tilting of the octahedral axes with respect to the c-axis is minimal [31]. 3.7. Band gap The energy band is a key feature reflecting the photocatalytic activity of a photocatalyst material [7]. UVevis absorption spectrum of PrFeO3 nanotubes (550 C/6 h) was tested (see Supporting Information Fig. S8). One can see that PrFeO3 nanotubes exhibit obvious absorption in visible light regions, which can mainly be attributed to the band gap electronic transition from the valence band to the conduction band (O 2p/Fe 3d) [7]. Based on the data of absorption spectra, the band gap can be calculated by using the formula (band gap ðeVÞ ¼ ð1240=lg Þ nm [15]; where lg is the absorption wavelength value obtained from absorbance spectrum).
Table 3 Refined atomic coordinate parameters of PrFeO3.
Fig. 3. (a) XRD patterns of electrospun nanofibers before and after calcination treatment, (b) representative experimental (crossed) and calculated (red solid line) X-ray diffraction profiles of PrFeO3 nanotubes (inset is the schematic views of the structure geometries along the c-direction). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Atom
x/a
y/b
z/c
U[Å2]
Pr1 Fe1 O1 O2
0.9946 (18) 0.00000 0.094 (9) 0.716 (6)
0.0408 (5) 0.50000 0.475 (5) 0.301 (5)
0.25000 0.00000 0.25000 0.040 (5)
0.0045 0.0078 0.029 0.022
(11) (18) (13) (10)
C. Qin et al. / Journal of Power Sources 285 (2015) 178e184
183
The band gap of the porous PrFeO3 nanotubes was 1.97 eV (lg ¼ 629 nm). This value is similar to that of nanocrystalline PrFeO3 synthesized by a combustion method (1.88 eV), template method (2.03 eV) and solegel method (2.08 eV) [15]. The absorption spans to the wavelength of 629 nm also indicates that PrFeO3 nanotubes could be used for visible-light photocatalysis [7]. 3.8. Photocatalytic activity To prove the photocatalytic activity in visible light, PrFeO3 nanotubes were placed in an aqueous solution containing RhB, and a desk lamp (60 W) with a UV filter was used as a light source to irradiate the solution. Fig. 4a shows the UVevis spectra taken during the photodegradation test. The peaks at 553 nm and 503 nm are the characteristic absorption of tetraethylated rhodamine and the absorption of fully de-ethylated RhB molecule [5]. Irradiation by the desk lamp, the solution color changed gradually from red to light green-yellow and finally colorless. The inset images show the color change of the RhB solution. After 30 h of irradiation, 90.6% of RhB was decolorized and the solution turned almost colorless, indicating that chromophoric structure was destroyed [5]. It was reported that photocatalysts in fibrous form were easier to separate from reaction solution than nanoparticles [17,32]. Here, PrFeO3 nanotubes were recycled by centrifuge for 5 min and its repeating usability was tested by decolorization of fresh RhB solutions with the used catalyst. As shown in Fig. 4b, in the 1st, the 2nd and the 3rd test cycles, the decolorization efficiency of 90.6%, 87.9% and 86.3% was achieved respectively. The results indicate that the PrFeO3 nanotubes can be used for multicycle photodegradation. The Supporting Information in Fig. S9 displays the XRD patterns of the PrFeO3 nanotubes before and after the three cycles of photodegradation. No obvious change could be found between the two curves, which indicated that the PrFeO3 nanotubes were not photo corroded during the photodegradation reaction. Fig. 4c shows the photo-degradation efficiency of RhB using different PrFeO3. 46.5% RhB is degraded by the nanotube sample in 6 h, while for the porous fiber and nanoparticle samples, the decolorization efficiency was 29.5% and 15.7%, respectively. And all the residual RhB aqueous were still in red. 12 h later, the degradation efficiency of 56.2%, 49.4% and 24.2% was achieved for the porous nanotube, porous nanofiber and nanoparticle samples, respectively. The residual RhB solutions were green-yellow, light red and red. 90.6% RhB was degraded by the nanotube sample in 30 h, while for the porous fiber and nanoparticle samples, 24.6% and 58.6% of RhB were still left, respectively. The nanotube sample exhibited the highest efficiency for its highest specific surface area [20] (see Table 1). The photodegradation efficiency of PrFeO3 nanotubes calcinated at different temperature was also tested (see Supporting Information Fig. S10). The efficiency decreased when the temperature increased from 550 to 700 C and the highest one was the sample calcined at 550 C for 6 h. This can be explained by that the sample (550 C/6 h) has the suitable crystallinity and specific surface area which are two important parameters determining photocatalytic activity. The sample (550 C/4 h) exhibited slightly lower efficiency than the sample (550 C/6 h) for poor crystallinity (see XRD spectrum in Fig. 3a) although it has the highest specific surface area. 4. Conclusions
Fig. 4. (a) UVevis spectral change of aqueous RhB solution at different photodegradation (PrFeO3 nanotubes, 550 C/6 h), (b) repeatability tests on the sample, and (c) photodegradation efficiency (h) of RhB solution in presence and absence of different PrFeO3 photocatalysts.
Perovskite-type PrFeO3 porous nanotubes with diameter 70e80 nm and wall thickness around 10 nm have been prepared successfully. The low temperature annealing treatment of precursor nanofibers was found to play a key role in determining the formation of porous PrFeO3 nanotubes. The porous tubes showed a
perovskite-type PrFeO3 crystal characteristic with a high absorption in the UVevisible light region and an energy band gap as low as 1.97 eV. When compared with porous PrFeO3 nanofibers and PrFeO3 particles, the PrFeO3 porous nanotubes showed higher
184
C. Qin et al. / Journal of Power Sources 285 (2015) 178e184
visible-light photocatalytic ability to degrade Rhodamine B in aqueous phase because of the increased specific surface areas and more active catalytic sites on the inner walls and outer surfaces. Acknowledgments The authors thank financial support from National Natural Science Foundation of China (No. 51273134), Jiangsu Provincial Natural Science Foundation of China (No. BK2012635) and Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD). Alfred Deakin Postdoctoral Fellowship awarded to the 1st author is also acknowledged. Appendix A. Supplementary data Supplementary data related to this article can be found online at http://dx.doi.org/10.1016/j.jpowsour.2015.03.096. References [1] S. Li, Y. Guo, L. Zhang, J. Wang, Y. Li, Y. Li, B. Wang, J. Power Sources 252 (2014) 21e27. [2] M. Basu, N. Garg, A.K. Ganguli, J. Mater. Chem. A 2 (2014) 7517e7525. [3] S.A. Ansari, M.M. Khan, M.O. Ansari, J. Lee, M.H. Cho, RSC Adv. 4 (2014) 26013e26021. [4] Y. Bi, S. Ouyang, J. Cao, J. Ye, Phys. Chem. Chem. Phys. 13 (2011) 10071e10075. [5] H. Fu, C. Pan, W. Yao, Y. Zhu, J. Phys. Chem. B 109 (2005) 22432e22439. [6] Y. Cheng, J. Chen, X. Yan, Z. Zheng, Q. Xue, RSC Adv. 3 (2013) 20606e20612. [7] S. Thirumalairajan, K. Girija, I. Ganesh, D. Mangalaraj, C. Viswanathan, A. Balamurugan, N. Ponpandian, Chem. Eng. J. 209 (2012) 420e428. [8] P. Tang, Y. Tong, H. Chen, F. Cao, G. Pan, Curr. Appl. Phys. 13 (2013) 340e343.
[9] K.M. Parida, K.H. Reddy, S. Martha, D.P. Das, N. Biswal, Int. J. Hydrogen Energy 35 (2010) 12161e12168. [10] S.N. Tijare, M.V. Joshi, P.S. Padole, P.A. Mangrulkar, S.S. Rayalu, N.K. Labhsetwar, Int. J. Hydrogen Energy 37 (2012) 10451e10456. [11] J. Ding, X. Lü, H. Shu, J. Xie, H. Zhang, Mater. Sci. Eng. B 171 (2010) 31e34. [12] L. Ju, Z. Chen, L. Fang, W. Dong, F. Zheng, M. Shen, J. Am. Ceram. Soc. 94 (2011) 3418e3424. [13] H. Yang, J.X. Zhang, G.J. Lin, T. Xian, J.L. Jiang, Adv. Powder Technol. 24 (2013) 242e245. [14] L. Li, X. Wang, Y. Lan, W. Gu, S. Zhang, Ind. Eng. Chem. Res. 52 (2013) 9130e9136. [15] S.N. Tijare, S. Bakardjieva, J. Subrt, M.V. Joshi, S.S. Rayalu, S. Hishita, N. Labhsetwar, J. Chem. Sci. 126 (2014) 517e525. [16] G-J. Lin, H. Yang, T. Xian, Chem. J. Chin. U 33 (2012) 1565e1571. [17] B. Dong, Z. Li, Z. Li, X. Xu, M. Song, W. Zheng, C. Wang, S.S. Al-Deyab, M. ElNewehy, J. Am. Ceram. Soc. 93 (2010) 3587e3590. [18] D. Hou, X. Hu, Y. Wen, B. Shan, P. Hu, X. Xiong, Y. Qiao, Y. Huang, Phys. Chem. Chem. Phys. 15 (2013) 20698e20705. [19] X. Li, C. Tang, M. Ai, L. Dong, Z. Xu, Chem. Mater. 22 (2010) 4879e4889. [20] S. Mohan, B. Subramanian, RSC Adv. 3 (2013) 23737e23744. [21] B. Lu, C. Zhu, Z. Zhang, W. Lan, E. Xie, J. Mater. Chem. 22 (2012) 1375e1379. [22] Y. Cheng, W. Huang, Y. Zhang, L. Zhu, Y. Liu, X. Fan, X. Cao, CrystEngComm 12 (2010) 2256e2260. [23] S. Zhan, D. Chen, X. Jiao, C. Tao, J. Phys. Chem. B 110 (2006) 11199e11204. [24] Z. Dong, S.J. Kennedy, Y. Wu, J. Power Sources 196 (2011) 4886e4904. [25] C.-Q. Chen, W. Li, C.-Y. Cao, W.-G. Song, J. Mater. Chem. 20 (2010) 6968e6974. [26] Z. Li, X. Wang, T. Lin, J. Mater. Chem. A 2 (2014) 13655e13660. [27] J.P. Cheng, B.B. Wang, M.G. Zhao, F. Liu, X.B. Zhang, Sens. Actuat. B: Chem. 190 (2014) 78e85. [28] S. Thirumalairajan, K. Girija, N.Y. Hebalkar, D. Mangalaraj, C. Viswanathan, N. Ponpandian, RSC Adv. 3 (2013) 7549e7561. [29] S.K. Pandey, R. Bindu, P. Bhatt, S.M. Chaudhari, A.V. Pimpale, Phys. B: Condens. Matter 365 (2005) 47e54. [30] M. Popa, J.M.C. Moreno, J. Alloy Compd. 509 (2011) 4108e4116. ndez-García, G. Colo n, Chem. Rev. 112 (2011) [31] A. Kubacka, M. Ferna 1555e1614. [32] D. Lin, H. Wu, R. Zhang, W. Pan, Chem. Mater. 21 (2009) 3479e3484.