An investigation of ZrO2: V nano materials

JOURNAL OF SCIENCE OF HNUE DOI: 10.18173/2354-1059.2015-0032 Mathematical and Physical Sci., 2015, Vol. 60, No. 7, pp. 55-61 This paper is available online at AN INVESTIGATION OF ZrO2:V NANOMATERIALS Nguyen Minh Thuy1, Le Thi Hong Hai2, Pham Thi Minh Thao2, Tran Thi May1, Nguyen Phuong Lien1 and Pham Tien Lam3 1Faculty of Physics, Hanoi National University of Education 2Faculty of Chemistry, Hanoi National University of Education 3Japan Advanced Institute of Science and Technology, Japan Abstract. ZrO2 nanopowders were synthesized using the hydrothermal method in an acid or base (alkaline) environment. The influence of preparation condition on the morphology, crystall phases and particle size in the sample has been investigated in order to obtain highly visible catalytic activity of ZrO2:V nanocrystals. The ZrO2:V nanoparticles have a monoclinic phase with average crystal size of 10 - 20 nm and they exhibit long tailed absorption in visible light. FTIR results show the existence of surface OH-groups on the sample, indicating capability of the ZrO2:V catalyst for biodiesel production. The catalytic activity of ZrO2:V nanocrystals examined while undergoingODH n-Butane reaction shows that Vanadium ions play an important role in the active catalytic centers in the reaction. The structural calculations such as band structure, density of states and optical properties were carried out under the framework of the Density Functional Theory (DFT). A theoretical analysis by computer simulation is expected to clarify the doping effects in detail and compare well with experimental results. Keywords: Zirconium, absorption, catalyst, density functional theory. 1. Introduction Zirconium ZrO2 is an important ceramic materials that possesses excellent thermal, dielectric, mechanical, chemical and biocompatibility properties [1, 2]. Zirconium is also a useful catalysis or an important support material for catalysis, having acidic and alkaline properties [1, 3]. In recent decades, ZrO2 mechanical properties such as elasticity, structural vibration and thermoelasticity) have been widely investigated [2-4]. Zirconium nano-particles will be widely applied in high-performance structural engineering ceramics and catalyst industries [1, 2]. Many methods have been used to obtain ZrO2 nano-particles or superfine powders, including vapor phase hydrolysis, the sol-gel process, the hydrothermal process and the combustion method or reverse micelles. Due to the wide band gap of zirconia (4 - 6 eV) it exhibits poor photocatalytic activities. However, heterogeneous ion doping adjusts the band gap and can make possible visible light absorption of ZrO2 [4]. In this work, ZrO2 and ZrO2:V nanopowders were synthesized using the hydrothermal method. We investigated the influence of the preparation condition on the morphology, crystal Received October 27, 2015. Accepted November 30, 2015. Contact Nguyen Minh Thuy, e-mail address: thuynm@hnue.edu.vn 55 Nguyen Minh Thuy, Le Thi Hong Hai, Pham Thi Minh Thao and Tran Thi May phases and particle size in the sample in order to obtain high visible photoactivity of ZrO2:V nanocrystals. A lack of understanding of the electronic structure and impurities existent in ZrO2:V nanocrystals leads to continued debate. A theoretical analysis by computer simulation is expected to clarify the doping effects in detail [5]. With this aim, the plane-wave ultrasoft pseudopotentials method within the framework of the density functional theory (DFT) [6, 7] has been adopted in this work. At the lattice parameters of supercells where ion doping occurred, electronic structures and optical properties of the optimized supercells were calculated. Our calculation results have been compared with experimental results and other calculated results. Based on these results, the mechanism of ZrO2 doped with ion has been discussed. 2. Content 2.1. Experimental and calculation method V-doped ZrO2 was synthesized by using ZrOCl2, citric acid (CA), NH4NO3, NH3 10%, V2O5/NaOH 2 M or HCl 2 M and C2H5OH. All chemicals were of analytical reagent grade. Pure Zirconium powder was prepared using the hydrothermal method and the following procedure: First, ZrOCl2 was added drop by drop to CA and NH4NO3 solvent (with V2O5/NaOH or HCl for the doping process), which was continuously stirred for 20 minutes Then the mixture was transferred into an autoclave and kept at 190 ◦C for 24 h. After centrifugation and washing, the samples were air dried at 50 ◦C. A series of V-doped ZrO2 hydrosol were prepared by changing the V/Ti ratio. The V doping contents are up to 3.2 molar%, examined by EDX measurement. The structure and crystalline characters were analyzed by X-ray diffraction (XRD) SIEMENS D5005. Absorption measurements were obtained with a JASCO V-670 spectrometer. Nitrogen adsorption-desorption isotherms and the Brunauer-Emmett-Teller (BET) surface areas were collected at 77 K using Micromeritics equipment. Calculations of total energy and electronic structure were carried out using the CASTEP package within the framework of DFT. The Perdew-Burke-Ernzerhof (PBE) parameterization of the generalized gradient approximation (GGA) [8] was adopted for the exchange-correlation potential. For all atoms, electron-core interactions are described by ultrasoft pseudopotentials. A cutoff energy of 380 eV and a regular Monkhorst-Pack grid of 10× 10× 10 k-points were adopted for the Brillouin zone sample. The implementation of the DFT method includes total energy and atomic-force calculations, which allow structure optimization [9, 10]. The optimized structures for the set unit cell volume (V) and the lattice constant were decided when the total energy and the force on each atom were minimized. All results in the study were obtained under this condition set. A Monkhorst-Pack scheme k-points grid sampling was used to create a irreducible Brillouin zone with kinetic energy cutoffs of 380.0 eV, the k-grid being 10× 10× 10,∆k = 0.02 (1/A), smearing being 10−6 ÷ 10−5 eV/atom. The primitive unit cell of ZrO2 is in the cubic and monoclinic structures. In the doping model calculations, a 96-atom monoclinic 2 × 2× 2 supercell (32 zircon and 64 oxygen atoms) was used in the construction of the V-doped structures (≈ 3.125%). We consider two situations: V-substitution for the Zr atom and for the V-interstitial atom. 56 An investigation of ZrO2:V nano materials 2.2. Result and discussion 2.2.1. Properties of the samples Figure 1 shows the X-ray diffraction patterns of pure ZrO2 (a-curve) and Vanadium doped ZrO2 with the V concentration up to 3% (b, c and d curves). For all samples, strong characteristic peaks of monoclinic ZrO2 are observed and assigned to corresponding crystal planes in the patterns (No 00-007-0343). Detailed analysis of the main characteristic peaks of the ZrO2 monoclinic (Figure 1, right) shows that the XRD main peak (named (-111) at 28.22◦) moved slightly toward the 2 Theta-increasing in area with increased Vanadium doping content, which can be explained by the V-incorporation into the ZrO2 host lattice. In all samples the width of the peak was broadened which indicates nanoparticle size. According to Scherer’s equation, the particle’s size is in range 10 - 20 nm, in good agreement with the below results. Figure 1. The XRD patterns of ZrO2 (a) and V doped ZrO2 (b, c, d) - left; the detail of (-111) peak-right Absorption spectra of ZrO2 (a), ZrO2:V (b, c, d) and V2O5 (e) are shown in Figure 2, left. The forbidden band gap energy of the ZrO2 and V-doped ZrO2 samples was calculated from optical absorption spectrum assuming indirect allowed transitions close to the fundamental absorption, using the following relationship [11]: αhν = A(hν − Eg)2 where hν is the photon energy, A is a photon energy independent parameter and Eg is the optical band gap energy. The calculated results are presented in Figure 2, right, and the Eg values are shown in Table 1. Pure ZrO2 (curve a) exhibited an absorption edge at around 5.06 eV (230 - 240 nm), which corresponds to the band gap of the monoclinic type of ZrO2 [12]. V-doped ZrO2 exhibited a red shift of the band edge and long tailed absorption above 340 nm. Compared with the spectrum of undoped ZrO2 (curve a), the band edges shifted systematically towards the smaller energy in the doped samples. V dopants in ZrO2 host were determined to preferentially substitute for Zr4+ ions [3], which can be confirmed by our modeling calculation later. The tailing of the absorption band in the doped samples can be assigned to the charge-transfer transition from the d orbital of V4+ to the conduction band of ZrO2. The absorption band around 3.0 - 4.5eV (340 nm) relates to the V doping effect [12]. The doping ZrO2:V material absorbed is up to 340 - 400 nm so it may lead to visible photocatalytic activity. 57 Nguyen Minh Thuy, Le Thi Hong Hai, Pham Thi Minh Thao and Tran Thi May Table 1. Band gap value of ZrO2 monoclinic and ZrO2:V Sample/%V Band gap value (eV) Exp Calc ZrO2 5.06 4.26 3.125 4.90 4.10 6.25 - 4.05 12.5 - 3.94 Figure 2. UV-Vis absorption of ZrO2 (a) and V-doped ZrO2 (b, c, d) The specific surface area of the materials has an impact on photocatalytic reaction’s performance. It can be investigated using the BET data, as the specific surface area of all samples is in correlation with particle size. Figure 3 shows the representative nitrogen adsorption and desorption isotherms and pore size distribution curve of ZrO2:V nanoparticles. The materials have a characteristic porous structure. The pore size distribution curve (Figure 3, right) shows that this sample has pores that are about 17 nm in size. The specific surface area is 23.9 m2/g based on the information in Figure 3, right. The catalytic behavior of the samples was examined using ODH reaction of n-butane. Pure ZrO2 was inactive. For the ZrO2:V nanoparticles, the buta-1,3-diene (or C4) were detected. This result indicates that ZrO2:V nanoparticles can be used in the Biodiesel production process [4]. Figure 3. Nitrogen adsorption and desorption isotherms (left) and pore size distribution curve (right) of ZrO2:V 58 An investigation of ZrO2:V nano materials 2.2.2. The calculation model In order to understand the doping ion role in the enhanced photocatalyst of ZrO2/V, we perform density functional theory (DFT) calculations. The core electrons were replaced by ultrasoft core potentials. Valence atomic configurations were 4s24p64d35s1 for Zr, 2s22p4 for O, and 3s23p63d54s2 for V atoms. Both the lattice parameter and the atomic position were optimized for pure and defective structure. The optimized lattice constants were calculated with different virtual functionals such as LDA, GGA + PBE, GGA + RPBE and GGA + PW91. The results show that the GGA structural results have been somewhat overestimated in comparison with experimental values, whereas the LDA results have been underestimated. From the comparison we adopt the optimization geometry of the host crystal obtained using the GGA + PBE functional. Optimized lattice constants are a = 5.208; b = 5.270 and c = 5.374 for the monoclinic phase, which are in good agreement with experimental values [4] and other calculations [13, 14]. The calculated band structures of monoclinic ZrO2 are shown in Figure 4. One can see that the top of the valence band and the bottom of the conduction band are at different points of the Brillouin zone. The compound has indirect band gap, the value is 4.2 eV, which is smaller than experimental data due to the well-known underestimation of conduction band state energies in the DFT model [6-8]. This result is in good agreement with other calculations [13, 14]. Figure 4. The band structures of monoclinic pure ZrO2 (left) and 3.125% vanadium doped ZrO2 (right) The composition of calculated energy bands can be resolved with the help of projected density of states (PDOS) and total density of states (DOS) diagrams (Figure 5). The valence band (VB) consists mainly of O 2p states and a small number of Zr 4d states while the conductance band (CB) consists of Zr 4d states with a small number of O 2p. The Fermi energy, being 0 eV on the energy axis, lies in the valence band. In the doped sample, the VB shifts towards a lower energy and an isolated V 3d state is localized below the bottom of the CB of the host ZrO2. The results are in good agreement with the absorption measurements and are similar to that in [14]. 59 Nguyen Minh Thuy, Le Thi Hong Hai, Pham Thi Minh Thao and Tran Thi May These gap states could be the reason of the narrowing of the band gap of the ZrO2:V materials and, if so, their visible light activity could be improved. Figure 5. DOS and PDOS of monoclinic pure ZrO2 (left) and the 3.125% vanadium doped ZrO2 (right) Figure 6 shows an absorption coefficient of ZrO2:V calculated with a GGA + PBE functional. The major optical absorption edge shifts to a longer wavelength when V- doping content increases. This agrees with experimental data (Figure 2). The Eg values, shown in Table 1, are smaller than the experimental data due to the well-known underestimation of the conduction band state energies in DFT calculations. The result is in good agreement with other calculations [13, 14]. The V doping induces a weak absorption edge below the major optical band gap, which results from the intermediate states and relates to the visible-light responses of this material. Figure 6. An abs. coefficient for pure ZrO2 and for ZrO2:V doped with 3.125, 6.25 and 12.5% vanadium 60 An investigation of ZrO2:V nano materials 3. Conclusion ZrO2 and ZrO2:V nanocrystals were successfully prepared using hydrothermal methods. All obtained samples were single monoclinic crystal phase. Average crystal size was about 10 - 20 nm. Pure ZrO2 exhibited an absorption edge at around 5.06 eV. The vanadium doped samples exhibited long tailed absorption in visible light above 340 nm. The DFT calculations of electronic structure, DOS and optical properties could explain the UV-Vis light responses of the materials. For the doped sample, VB shifts toward a lower energy, and isolated V 3d state is localized below the bottom of the CB of the host ZrO2. This result is in good agreement with other calculations. Acknowledgements. The authors would like to thank the Hanoi National University of Education. This work was supported by Ministry of Education and Training (MOET) Grant No. B2014-17-46. REFERENCES [1] B. M. Weckhuysen and D. E. Keller, 2003. Catalysis Today. Vol. 78, pp. 25-46. [2] R. G. Luthardt, M. Holzhuter, O. Sandkuhl, V. Herold, J. D. Schnapp, E. Kuhlish, and M. H. Walter, 2002. J. Dent. Res. 91, pp. 487-491. [3] J. Liu. Z. Zhao and C. M. Xu, 2005. Acta Phys. Chim. Sin., 21, N2, p. 156. [4] Delia Gazzoli, Sergio De Rossi, Giovanni Ferraris, Giorgio Mattei, Roberto Spinicci, Mario Valigi, 2009. Journal of Molecular Catalysis A: Chemical 310, pp. 17- 23. [5] Yi. Zhang, 2014. J. Zhang, Materials Today: Proceedings 1, pp. 44-55. [6] Perdew J. P., Wang Y., 1992. Phys. Rev. B 45, pp. 13244-13249. [7] Perdew, J. P., Zunger, A., 1981. Phys. Rev. B, 23, pp. 5048-5079. [8] Perdew J. P., Burke K., Ernzerhof M., 1996. Phys. Rev. Lett., 77, pp. 3865-3868. [9] Nguyen Minh Thuy, Duong Quoc Van and Le Thi Hong Hai, 2012. J. Nanomaterials and Nanotechnology, Vol. 2, Art 14:2012, Intechopen.com. [10] Minh Thuy Nguyen, Cao Khang Nguyen, Thi Mai Phuong Vu, Quoc Van Duong, Tien Lam Pham and Tien Cuong Nguyen, 2014. Adv. Nat. Sci: Nanosci. Nanotechnol. 5 045018. [11] E. Rosencher and Borge Vinter, Optoelectronics, 2002. Cambridge University Press, Frontmatter. [12] T. Ito, M. Maeda, K. Nakamura, H. Kato and Y. Ohki, 2005. J. Appl. Phys. 97, 054104. [13] Giuseppe Fadda, Luciano Colombo and Giovanni Zanzotto, 2009. Phys. Rev. B 79, 214102. [14] G. Jomard, T. Petit, A. Pasturel L. Magaud, G. Kresse and J. Hafner, 1999. Phys. Rev. B 59, Vol. 6, 4044. 61