Synthesis, growth mechanism and room-temperature blue luminescence emission of uniform WO 3 nanosheets with W as starting material Jinmin Wang, Pooi See Lee à , Jan Ma School of Materials Science and Engineering, Nanyang Technological University, Nanyang Avenue, Singapore 639798, Singapore article info Article history: Received 21 July 2008 Received in revised form 24 October 2008 Accepted 5 November 2008 Communicated by K. Nakajima Available online 13 November 2008 PACS: 68.70.+w 81.05.Dz 81.10.Dn 81.20.Ka Keywords: A1. Crystal morphology A2. Growth from solutions B1. Nanomaterials B1. Oxides abstract Uniform single-crystalline tungsten oxide (WO 3 ) nanosheets have been synthesized in a large-scale with W powders as starting material. The results from the thermal stability meas urements show that the as-synthesized WO 3 nanosheets are anhydrous. Their thickness and length are $80 and $500 nm, respectively. They exhibit blue luminescence emissions at $431, 486 and 497 nm, UV emissions at $362 and 398 nm. The blue emissions are resulted from the band–band indirect transition and the UV emissions should be attributed to the defect states of WO 3 . The growth mechanism of the two- dimensional WO 3 nanosheets is discussed. & 2008 Elsevier Ltd All rights reserved. 1. Introduction Blue luminescence emission has attracted much attention due to the applications in short-wavelength laser [1], light-emitting diode (LED) [2] and white light source [3]. A number of blue- emission semiconductor materials have been studied, such as GaAs, GaN, ZnSe [4–7]. These materials are direct band-gap semiconductors which readily emit luminescence. On the other hand, tungsten oxide (WO 3 ) is an indirect band-gap semiconduc- tor [8], which has been extensively studied due to their applications in electrochromic [9], photocatalytic [10] and gas sensing materials [11]. Less attention has been paid to the luminescence properties of WO 3 because of the low emission efficiency in conventional indirect band-gap semiconductors. However, recently, much progress has been realized for the luminescence of WO 3 . Manfredi et al. [12] reported the light emission in WO 3 thin films, at which the excitation temperature is at liquid nitrogen temperature. Niederberger et al. [13–15] realized the room-temperature blue emission of WO 3 nanoparti- cles in ethanol solution. Feng et al. [8] also reported the room- temperature strong photoluminescence (PL) of WO 3 nanoparticles and W 18 O 49 nanowires. It is believed that the particle size, morphology and quantum-confinement effect played an impor- tant role for the room-temperature luminescence emission [16]. Recently, a great deal of efforts has been focused on the morphology control of all kinds of nanostructures due to their morphology dependent properties [17–23]. Nanosheets, one of two-dimensional (2D) nanostructures with one distinct thin thickness, have many special potential applications in electrical, optical, photochemistry, sensors and ion-exchange properties [24]. Relative to the comprehensive investigations in zero- and one-dimensional (1D) nanostructures, 2D nanostructures are almost neglected for the last decade. One important reason is the synthesis of single-crystalline 2D nanostructures is more difficult to control. However, much progress has been achieved recently for the synthesis of ZnO, TiO 2 ,Fe 3 O 4 ,Mn 3 O 4 , CoO, Ga 2 O 3 and complex hydroxide nanosheets [24–31]. It is well known that tungsten oxides contain many non-stoichiometric sub-oxides (WO 3Àx ). Moreover, some hydrates often exist in the products from a wet chemical route for the synthesis of WO 3 , making it difficult to synthesize stoichiometric WO 3 nanosheets. Only a few research groups reported the synthesis of WO 3 and tungsten oxide hydrates (WO 3 Á xH 2 O) nanosheets. Polleux et al. [32] successfully synthesized tungsten oxide hydrate (WO 3 Á H 2 O) nanoplatelets by ARTICLE IN PRESS Contents lists available at ScienceDirect journal homepage: www.elsevi er.com/locate/jcrysgro Journal of Crystal Growth 0022-0248/$ - see front matter & 2008 Elsevier Ltd All rights reserved. doi:10.1016/j.jcrysgro.2008.11.016 à Corresponding author. Tel.: +65 67906661; fax: +65 67909081. E-mail address: pslee@ntu.edu.sg (P.S. Lee). Journal of Crystal Growth 311 (2009) 316–319 heating the system of tungsten chloride (WCl 6 )-benzyl alcohol. Considering the poor thermal stability of WO 3 Á H 2 O as it can decompose and release water molecules at higher temperatures, anhydrous WO 3 will be more stable for practical applications. Kim et al. [33] reported a similar system of WCl 6 in ethanol to synthesize WO 3 nanosheets. Recently, Zhang et al. [34] synthe- sized ultrathin WO 3 nanodisks using poly(ethylene glycol) (PEG) as a surface modulator. However, the luminescence properties of WO 3 nanosheets were not studied and only the last paper proposed a possible growth mechanism for the WO 3 nanodisks. Moreover, when WCl 6 was used as the starting material, the reaction medium must be organic solvent because WCl 6 will quite strongly hydrolyze to release hydrogen chloride in aqueous solution. Besides, it is expensive to use WCl 6 as the starting material. So an aqueous approach using common tungsten source must be developed for the large-scale synthesis of WO 3 nanosheets. Obviously, metal tungsten (W) is the ideal tungsten source for the preparation of WO 3 . However, up to now, no one has reported the synthesis of WO 3 nanosheets using W as the starting material. Herein, we report a much more economical approach to the large-scale synthesis of uniform stoichiometric and anhydrous WO 3 nanosheets from W powder and a successful demonstration for the room-temperature blue emissions from the resultant WO 3 nanosheets. The blue emissions were attributed to arise from the band–band indirect transition of WO 3 . And a new growth mechanism for the as-synthesized WO 3 nanosheets was proposed. 2. Experimental procedure The precursor was prepared by oxidation of W using hydrogen peroxide (H 2 O 2 ) [35,36]. In a typical synthesis, 6.5 g of W powder was dissolved into a mixed solution of 40.0 mL of H 2 O 2 and 4.0 mL of H 2 O with an ice-water bath and stirring. After filtration, a light yellow solution was obtained. The solution was refluxed at 55 1C for 6 h, then yellow concentrated sol was formed. After aging at room temperature, yellow precipitate was separated out and used as the precursor. A total of 0.4 g of precursor was added into 19.0 mL of de-ionized water to form a suspension and 3 M HCl was dropped into the suspension until its pH value is up to 1.7. Then, the suspension was transferred into a Teflon-lined autoclave with a capacity of 45 mL. The autoclave was placed into an oven and heated at 180 1C for 24h. After cooling down to room temperature, yellow product was obtained. The phase of the product was identified by X-ray powder diffraction (XRD), using Cu K a ( l =0.15406 nm) radiation in a 2 y range from 101 to 801 at room temperature. The morphologies of the as-synthesized WO 3 nanosheets were characterized by field- emission scanning electron microscopy (FESEM). The FESEM sample was prepared by coating Pt using a sputtering machine at a beam current of 20 mA for 45 s. Transmission electron microscopy (TEM) and high-resolution TEM (HRTEM) images, selected area electron diffraction (SAED) pattern of the WO 3 nanosheets were obtained using an accelerating voltage of 200 kV. The thermal stabilities of the precursor and the product are examined by thermal gravimetric analysis (TGA) in air. The PL properties of the as-synthesized WO 3 nanosheets were measured on a fluorescence spectrophotometer with a Xe lamp as the excitation source at room temperature. 3. Results and discussion The XRD patterns of the precursor and the product are shown in Fig. 1a. All the XRD peaks of the precursor can be identified as tungsten oxide hydrate (WO 3 Á H 2 O) (JCPDS 18-1419). Compared to the XRD pattern of the precursor, significant differences can be found from the XRD pattern of the product, showing a chemical reaction has occurred in the process of hydrothermal treatment. All of the XRD peaks of the product can be indexed to the orthorhombic structure of WO 3 (JCPDS 71-0131). No non- stoichiometric tungsten oxides (WO 3Àx ) and tungsten oxide hydrates (WO 3 Á xH 2 O) were detected, indicating pure orthor- hombic WO 3 has been obtained. Strong and sharp diffraction peaks also indicate good crystallinity of the hydrothermal product. Fig. 1b shows TGA curves of the precursor and the product. It can be seen that the precursor has an obvious weight loss within 150–200 1C, corresponding to the decomposition of WO 3 Á H 2 O and formation of WO 3 . In contrary, the TGA curve of the product is almost a straight line, which implies the product do not contain any hydrate. This result from thermal analysis is well consistent with the XRD result. Fig. 2a and b shows the FESEM image and TEM image of the as-synthesized WO 3 nanosheets. Uniform WO 3 nanosheets with a thickness of $80 nm and length of $500 nm can be observed. Further structural characterizations were carried out by HRTEM. The clear crystal lattices show that the as-synthesized WO 3 nanosheets are single crystals. The calculated lattice spaces are 0.385 and 0.376nm in the 2D plane of a single nanosheet, which correspond to the plane distances of (0 0 2) and (0 2 0) planes, respectively. And the angle between the (0 0 2) and (0 2 0) plane is 901. These crystal parameters are well consistent with that from standard XRD data. It can be regarded that the WO 3 crystals grew ARTICLE IN PRESS (001) (020) (200) (220) (021) (111) 10 100 98 96 product precursor 94 92 Weight loss (%) Intensity/(a.u.) 20 30 2 /(° ° ) Temperature ( ° C) 40 50 60 70 80 50 100 150 200 250 300 350 400 (121) (221) (002) (400) WO 3 WO 3 . H 2 O Fig. 1. (a) XRD patterns and (b) TGA curves of the precursor and the product (showing no weight loss is detected in the product). J. Wang et al. / Journal of Crystal Growth 311 (2009) 316–319 31 7 along two perpendicular directions, [0 0 2] and [0 2 0] to form 2D nanosheets. The formation process of WO 3 nanosheets from W can be divided into three parts: the formation of precursor, the formation of WO 3 by decomposition of WO 3 Á H 2 O and growth of WO 3 crystal nucleus. Peroxy-tungstates were formed by the action of H 2 O 2 on W [35,36]. After ageing, the precursor WO 3 Á H 2 O precipitated from the solution, which has been proved by XRD data (Fig. 1a). The results from thermal analysis show the precursor WO 3 Á H 2 O can decompose into anhydrous WO 3 at lower temperature than the applied hydrothermal temperature. So WO 3 can be easily formed in our experimental conditions. However, the subsequent crystal growth is dominant for the formation of WO 3 nanosheets. According to the existing steps of the WO 3 nanosheets (Fig. 2a, b and d), it is believed that the obtained WO 3 nanosheets grow by step-growth mechanism which resulted from the periodic bond chain (PBC) theory developed by Kossel, Stranki and Volmer [37]. According to the step-growth theory [37], an atom adsorbed onto a facet will diffuse randomly on the surface until it reaches an energetically favorable site and will subsequently be incorporated into the crystal structure by generating chemical bonds. This growth process will reduce the total energy of the atom and the crystal. If more chemical bonds can be generated at some site, that site will be the preferred growth site. Compared with a terrace (smooth surface), a corner of a given crystal can provide more unsatisfied bonds to generate new chemical bonds with an incorporated atom. Thus, the crystal will be in a thermodynami- cally stable state. Why did the WO 3 crystal nucleus grow into 2D nanosheets instead of 3D crystals? This is due to the distinct growth rate of different crystal facets with different surface energy. It is well known that the facets have different atomic densities and unsatisfied bonds, resulting in variations of surface energy. The facet with a larger surface area has a smaller surface atom density, which results in a lower surface energy. For orthorhombic WO 3 with a=7.341, b=7.570 and c=7.754, the (2 0 0) facet has the largest surface area and lower surface energy, resulting in a lowest growth rate in the [2 0 0] direction. In contrary, the other directions of [0 0 2] and [0 2 0] have higher growth rate, resulting in the preferred growth along [0 02] and [0 2 0] directions. Hence the resultant shape of the crystals is a 2D nanosheet. This mechanism differs from WO 3 nanorods or nanodisks growth [23,34] which emphasizes the use of surface modulators or structure-directing agents. For the growth of WO 3 nanorods, some capping agents (Cl À ions) cap some facets of WO 3 crystal nuclei, resulting in slow growth rates of these facets and one fast growth rate of a special direction (c-axis) [23]. For the growth of WO 3 nanodisks [34], it is believed that the formation of WO 3 nanodisks is driven by the preferential adsorption of poly(ethylene glycol) (PEG—10 0 0 0) onto the (0 1 0) crystal facets of WO 3 , thereby inhibiting crystallographic growth. However, the growth process of the as-synthesized WO 3 nanosheets does not dependent on any surface modulators or structure-directing agents, which belongs to facet growth. The PL properties of the as-synthesized WO 3 nanosheets were measured using a Xe lamp as the excitation source at room temperature. Fig. 3 shows the excitation and emission spectrum. When the excitation wavelength is 315 nm (Fig. 3a), the emission peaks of the as-synthesized WO 3 nanosheets consist of two UV emissions at $362 and 398 nm and three blue luminescence emissions at $431 (2.88 eV), 486 (2.55 eV) and 497 nm (2.4 9 eV) (Fig. 3b). The blue emissions (2.88, 2.55, 2.49 eV) are in the range of reported band-gap energies of WO 3 [12,38,39], so the blue emissions should be attributed to the indirect band–band ARTICLE IN PRESS Fig. 2. (a) FESEM, (b) TEM and (c), (D) HRTEM images of the as-synthesized WO 3 nanosheets. 275 Wavelength (nm) Intensity (%) Intensity (%) Wavelength (nm) 300 325 350 400 450 500 Fig. 3. (a) Excitation spectrum and (b) room-temperature photoluminescence spectrum of the as-synthesized WO 3 nanosheets. J. Wang et al. / Journal of Crystal Growth 311 (2009) 316–319318 transition of WO 3 and the UV emissions should arised from the defect states of WO 3 . Niederberger et al. [13] also suggested that the blue emission resulted from band–band transition of WO 3 . Recently, Zhao et al. [40] also found similar PL properties in WO 3Àx nanowire networks and attributed it to the above- mentioned explanation by measuring the changes of emission peaks with the changing excitation wavelengths. 4. Conclusions In summary, uniform single-crystalline WO 3 nanosheets have been synthesized in a large-scale with W powders as starting material by a facile hydrothermal process. The thermal stabilities of the precursor and the as-synthesized WO 3 nanosheets were studied. The results show that the WO 3 nanosheets are stoichio- metric and anhydrous whereas the precursor contains hydrates. The developed novel process for the synthesis of WO 3 nanosheets is much more economical than the previous methods. The growth mechanism of the 2D WO 3 nanosheets is proposed. The as- synthesized WO 3 nanosheets successfully exhibit blue lumines- cence emissions at $431, 486 and 497 nm, UV emissions at $362 and 398 nm. The blue emissions are resulted from the band–band indirect transition and the UV emissions should be attributed to the defect states of WO 3 . Acknowledgement The authors thank Mr. Liap Tat Su for the assistance in PL measurements. References [1] Y L. Lai, C P. Liu, Z Q. Chen, Appl. Phys. Lett. 86 (2005) 121915. [2] P.O. Anikeeva, J.E. Halpert, M.G. Bawendi, V. Bulovic, Nano Lett. 7 (2007) 2196. [3] K. Takahashi, N. Hirosaki, R J. Xie, M. Harada, K. Yoshimura, Y. Tomomura, Appl. Phys. Lett. 91 (2007) 091923. [4] G. Fasol, Science 272 (1996) 1751. [5] S. Nakamura, Science 281 (1999) 956. [6] M.A. Reshchikov, P. Visconti, H. Morkoc, Appl. Phys. Lett. 78 (2001) 177. [7] C C. Pan, C M. Lee, J W. Liu, G T. Chen, J I. Chyi, Appl. Phys. Lett. 84 (2004) 5249. [8] M. Feng, A.L. Pan, H.R. Zhang, Z.A. Li, F. Liu, H.W. Liu, D.X. Shi, B.S. Zou, H.J. Gao, Appl. Phys. Lett. 86 (2005) 141901. [9] C. Santato, M. Odziemkowski, M. Ulmann, J. Auustynski, J. Am. Chem. Soc. 123 (2001) 10639. [10] S H. Baeck, K S. Choi, T.F. Jaramillo, G.D. Stucky, E.W. McFarland, Adv. Mater. 15 (2003) 12698. [11] X.L. Li, T.J. Lou, X.M. Xiao, Y.D. Li, Inorg. Chem. 43 (20 04) 5442. [12] M. Manfredi, C. Paracchini, G.C. Salviati, G. Schianchi, Thin Solid Films 79 (1981) 161. [13] M. Niederberger, M.H. Bartl, G.D. Stucky, J. Am. Chem. Soc. 124 (2002) 13642. [14] K. Lee, W.S. Seo, J.T. Park, J. Am. Chem. Soc. 125 (2003) 3408. [15] T. Takagahara, K. Takeda, Phys. Rev. B 46 (1992) 15578. [16] A.N. Khold, V.L. Shaposhnikov, N. Sobolev, S. Ossicini, Phys. Rev. B 70 (2004) 035317. [17] F.L. Deepak, C.P. Vinod, K. Mukhopadhyay, A. Govindaraj, C.N.R. Rao, Chem. Phys. Lett. 353 (2002) 345. [18] G. Gundiah, A. Govindaraj, C.N.R. Rao, Chem. Phys. Lett. 351 (2002) 189. [19] C.N.R. Rao, F.L. Deepak, G. Gundiah, A. Govindaraj, Prog. Solid State Chem. 31 (2003) 5. [20] J.M. Wang, L. Gao, J. Mater. Chem. 13 (20 03) 2551. [21] J.M. Wang, L. Gao, J. Crystal Growth 262 (2004) 290. [22] C. Burda, X. Chen, R. Narayanan, M.A. EI-Sayed, Chem. Rev. 106 (2005) 1025. [23] J.M. Wang, E. Khoo, P.S. Lee, J. Ma, J. Phys. Chem. C 112 (2008) 14306. [24] L. Li, R. Ma, Y. Ebina, K. Fukuda, K. Takada, T. Sasaki, J. Am. Chem. Soc. 129 (2007) 8000. [25] J H. Park, H J. Choi, Y J. Choi, S H. Sohn, J G. Park, J. Mater. Chem. 14 (2004) 35. [26] T. Yui, Y. Mori, T. Tsuchino, T. Itoh, T. Hattori, Y. Fukushima, K. Takagi, Chem. Mater. 17 (2005) 206. [27] K.C. Chin, G.L. Chong, C.K. Poh, L.H. Van, C.H. Sow, J. Lin, A.T.S. Wee, J. Phys. Chem. C 111 (2007) 9136. [28] Y. Oaki, H. Imai, Angew. Chem. Int. Ed. 46 (2007) 4951. [29] W. Zhang, M. Han, Z. Jiang, Y. Song, Z. Xie, Z. Xu, L. Zheng, Chem. Phys. Chem. 8 (2007) 2091. [30] U.K. Gautam, S.R.C. Vivekchand, A. Govindaraj, C.N.R. Rao, Chem. Commun. (2005) 3995. [31] G. Gundiah, A. Govindaraj, C.N.R. Rao, Chem. Phys. Lett. 351 (2002) 189. [32] J. Polleux, N. Pinna, M. Antonietti, M. Niederberger, J. Am. Chem. Soc. 127 (2005) 15595. [33] H.G. Choi, Y.H. Jung, D.K. Kim, J. Am. Ceram. Soc. 88 (2005) 1684. [34] A. Wolcott, T.R. Kuykendall, W. Chen, S. Chen, J.Z. Zhang, J. Phys. Chem. B 110 (2006) 25288. [35] M. Deepa, A.K. Srivastava, S. Singh, S.A. Agnihotry, J. Mater. Res. 19 (2004) 2576. [36] E.A. Meulenkamp, J. Electrochem. Soc. 144 (1997) 1664. [37] G.Z. Cao, Nanostructures and Nanomaterials: Synthesis, Properties and Applications, Imperial College Press, London, 2004. [38] C.G. Granqvist, Handbook of Inorganic Electrochromic Materials, Elsevier, Amsterdam, 1995. [39] B. Yang, Y. Zhang, E. Drabarek, P.R.F. Barnes, V. Luca, Chem. Mater. 19 (2007) 5664. [40] J.Y. Luo, F.L. Zhao, L. Gong, H.J. Chen, J. Zhou, Z.L. Li, S.Z. Deng, N.S. Xu, Appl. Phys. Lett. 91 (2007) 093124. ARTICLE IN PRESS J. Wang et al. / Journal of Crystal Growth 311 (2009) 316–319 31 9 . Synthesis, growth mechanism and room-temperature blue luminescence emission of uniform WO 3 nanosheets with W as starting material Jinmin Wang, Pooi. The blue emissions were attributed to arise from the band–band indirect transition of WO 3 . And a new growth mechanism for the as- synthesized WO 3 nanosheets