- Email: [email protected]
Fabrication and photocatalytic activity of porous TiO2 nanowire microspheres by surfactant-mediated spray drying process
Materials Research Bulletin 44 (2009) 1070–1076
Contents lists available at ScienceDirect
Materials Research Bulletin journal homepage: www.elsevier.com/locate/matresbu
Fabrication and photocatalytic activity of porous TiO2 nanowire microspheres by surfactant-mediated spray drying process Xiwang Zhang a,*, Jia Hong Pan a, Alan Jianhong Du a, Jiawei Ng a, Darren D. Sun a,*, James O. Leckie b a b
School of Civil and Environmental Engineering, Nanyang Technological University, 639798 Singapore, Singapore Department of Civil and Environmental Engineering, Stanford University, Stanford, CA 94305-4020, USA
A R T I C L E I N F O
A B S T R A C T
Article history: Received 30 June 2008 Received in revised form 18 September 2008 Accepted 24 October 2008 Available online 5 November 2008
A novel, porous TiO2 nanowire microsphere with a diameter of 3–8 mm was successfully fabricated via spray drying of TiO2 nanowire suspension with the assistance of surfactant (F127). The products were characterized by FESEM, XRD and N2 adsorption–desorption analysis and results revealed that the resulting TiO2 nanowire microspheres possessed a hierarchically macro/mesoporous structure, as well as a high BET surface area of 38.2 m2/g. Systematic studies showed that the presence of surfactant in the suspension feed for spray drying was critical in the formation of porous microspheres. The structure of the fabricated microspheres depends on the nanowire concentration in the feed. The TiO2 nanowire microspheres exhibited significant photocatalytic degradation of Methylene blue (MB) as compared to commercial TiO2 nanoparticles (P25). It was also revealed that the microspheres have excellent stability on photocatalytic activity and mechanical strength, which are both crucial factors when considering reuse of these photocatalysts. ß 2008 Elsevier Ltd. All rights reserved.
Keywords: A. Nanostructures A. Semiconductors C. Electron microscopy D. Catalytic properties
1. Introduction Due to its unique physicochemical properties, TiO2 has received considerable attention in areas of environmental purification, solar cell and gas sensor, etc. [1,2]. Recently, one-dimensional (1D) nanostructured TiO2 has gained greater popularity due to its superior performances relative to conventional bulk materials, as a result of its large surface area and nanosize effect. Recently 1D TiO2 with various morphologies (wires, fibers and tubes) have been fabricated [3–5]. However, the practical environmental applications of these 1D TiO2 materials is a huge problem. They are too small to be separated and reclaimed by conventional separation methods. From an engineering point of view, besides superior photocatalytic activity, proper photocatalysts should be easily reclaimed for repeated usages. Otherwise, it would be deemed impractical because too much TiO2 catalysts would be expended. Additionally, the unreclaimed TiO2 might result in secondary pollution [6]. Micron-sized solid spherical TiO2 photocatalysts have been synthesized by some researchers [7–11] as an ideal candidate for
* Corresponding authors. Tel.: +65 6790 6273; fax: +65 6791 0676. E-mail addresses: [email protected] (X. Zhang), [email protected] (D.D. Sun). 0025-5408/$ – see front matter ß 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.materresbull.2008.10.017
easy separation and recovery, since they can settle down easily in aqueous suspensions by gravity. Unfortunately, a significant loss in contact area between the microspherical photocatalysts and UV light limits the photocatalytic efficiency. Grime and co-workers [12] and Wang et al. [13] proved that macroporous channels in TiO2 could serve as light-transfer paths for the distribution of photon energy, which would improve the photoactivity. More recently, template-assisted methodology has been developed to fabricate TiO2 microspheres with meso/macroporous framework. Sanchez’s group has prepared multi-scale porous submicron-sized TiO2 spheres by spray drying a Ti-sol containing supremolecular template (Pluronic F127) [14]. Okuyama’s group has fabricated brookite TiO2 multi-scale porous spheres by spray drying a suspension of brookite nanoparticles and polystyrene latex (PSL) particles [15]. In our previous work [16], we found that TiO2 (B) nanowire basketry-like microspheres can be synthesized by spray drying a suspension of nanowire and polythethylene glycol (PEG). In this paper, the method was continually studied to fabricate TiO2 nanowire microsphere with controllable structure with the assistance of surfactants. The formation mechanism of the TiO2 nanowire microspheres was studied by investigating the effects of different operating parameters. Thereafter, its activity and stability were evaluated by the batch photocatalytic oxidation of a model pollutant.
X. Zhang et al. / Materials Research Bulletin 44 (2009) 1070–1076
2. Experimental 2.1. Fabrication of the TiO2 nanowire microspheres The fabrication of nanowires was commonly referred in literatures [17–19]. In a typical synthesis, TiO2 powder (Degussa, P25) was mixed with 10 M NaOH solution in a Teflon-lined autoclave container and placed in the oven at 180 8C for 2 days to undergo a hydrothermal reaction. The white pulp-like product in the autoclave was washed with 0.1 M HCl, with the assistance of ultrasound. Subsequently, pH was neutralized by repeated washing with deionized water. The powdered nanowires were added into 0.1 wt% aqueous Pluronic F127 solution to obtain 8 g/L TiO2 suspension feed for spray dryer (EYELA SD-1000). As-synthesized TiO2 nanowire microspheres were constructed by spray drying the suspension feed. Finally, they were calcined at 300–600 8C for 2 h with a heating ramp of 2 8C/min to obtain the final nanowire microspheres. 2.2. Characterization of the TiO2 nanowire microspheres A JEOL 6340 field emission scanning electron microscopy (FESEM) and a JEOL 2010 transmission electron microscopy (TEM) were used to observe the morphologies of the TiO2 nanowire microspheres. The crystal structures and the phase compositions of the samples were identified using a Bruker AXS D8 Advance Xray diffractometer with monochromated high-intensity Cu Ka irradiation (l = 1.5406 A˚) at a scanning rate of 28/min. The N2 adsorption–desorption isotherms were obtained at liquid nitrogen temperature (77 K) using a Quantachrome Autosorb1 instrument. Before the measurement, the samples were outgassed under vacuum for 5 h at 150 8C. Specific surface area was calculated by
1071
the Brunaur–Emmett–Teller (BET) method using the adsorption data at the range from P/P0 = 0.05 to 0.35 just below the capillary condensation, and the pore diameter distribution curve was derived from the adsorption branch by the BJH method. A thermalgravimeter analyzer (TGA) was used to monitor the degradation of surfactant in the sample. Measurements were taken with a heating rate of 5 8C/min from 30 to 700 8C. 2.3. Evaluation of the photocatalytic activities of TiO2 nanowire microspheres Methylene blue (MB) was used as the model chemical to investigate the photocalytic activity of the TiO2 nanowire microspheres. An Upland 3SC9 Pen-ray lamp (254 nm) was immersed into the solution as the UV source. Air was pumped into the solution to mix for the catalysts and solution, as well as to induce oxygen into the system for oxidation. The aqueous system containing MB (20 mg/L, 500 mL) and TiO2 nanowire microspheres (0.5 g/L) was mixed in the dark for 30 min to attain the adsorption equilibrium of MB with the photocatalyst, before the UV lamp was switched on. Commercial TiO2 (Degussa P25) was adopted as the reference for comparison. The characteristic absorption at l = 670 nm was chosen to monitor the concentration of MB during the photodegration process. TOC (total organic carbon) was measured using a Shimadzu TOC-5000 analyzer. 3. Results and discussion FESEM images of the as-synthesized and calcined TiO2 samples at different magnifications are shown in Fig. 1. From the low magnification image (Fig. 1a) of the as-synthesized sample, it was found that microspheres, having 3–10 mm in diameter, were produced in vast amounts. The spherical structure of the TiO2
Fig. 1. FESEM images of (a and b) as-synthesized and (c and d) calcined nanowire microspheres at 600 8C. Nanowire and F127 concentration in suspension feed: 8 g/L and 0.1 wt%.
1072
X. Zhang et al. / Materials Research Bulletin 44 (2009) 1070–1076
photocatalysts retained without apparent decrease in diameter even calcined at 600 8C (Fig. 1c). A typical FESEM image of a single as-synthesized microsphere is presented in Fig. 1b, in which it can be easily seen that TiO2 nanowires were enwrapped by the surfactant to form a microsphere. After calcination the surfactants were completely removed and a porous structure was achieved in the microspheres (Fig. 1d). The crystal structures and the phase compositions of the nanowire microspheres calcined at different temperatures were identified by wide-angle XRD as shown in Fig. 2. The diffraction pattern of the as-synthesized nanowires is similar to that as expected for protonated titanate (H2Ti2O5) with lepidocrociterelated layered structure (JCPDS 47-0124). No additional diffraction peaks of other impurities, including the anatase or rutilephased TiO2, and NaCl were detected. These results are consistent with the previous reports [20,21]. Upon calcination at 300 8C, some XRD peaks for protonated titanate became weaker or disappeared, while some new diffraction peaks of TiO2 (B), (JCPDS 35-0088) appeared. The dehydration of protonated titanate and the phase transition process were accelerated by sintering the samples at higher temperatures. Upon calcination at 500 8C, the XRD peaks of protonated titanate completely disappeared and the resulting XRD peak positions match well with the diffraction of the TiO2 (B), indicating that the protonated titanates were fully dehydrated and converted into TiO2 (B). Further increase in calcination temperatures would result in phase transformation from TiO2 (B) to anatase (JCPDS 21-1272). At 550 8C, anatase phase was found to coexist with TiO2 (B). Pure anatase phase achieved when the sample was calcined at 600 8C. The texture of TiO2 nanowire microspheres was further understood by surface area analysis and related pore-size distribution. Fig. 3 shows the nitrogen adsorption–desorption isotherm of TiO2 calcined nanowire microsphere. BET surface area was determinated to be 38.2 m2 g 1, which is very close to that of TiO2 nanowire powder samples obtained without spray drying (39.6 m2 g 1). As shown in Fig. 1b, the interlaced nanowires constitute to the walls of the microspheres and create macropores with pore sizes ranging from 50 to 150 nm. Moreover, the insert of Fig. 3 shows the pore-size distribution plot of TiO2 nanowire microspheres exhibiting a mean pore diameter of 5 nm. Thus, the multi-scale macro/meso-porosity of the resulting TiO2 nanowire microspheres is high enough to prevent a decline in surface area values after spray drying and
Fig. 2. XRD patterns of the nanowire microspheres calcined at different temperature.
Fig. 3. Nitrogen adsorption and desorption isotherms and pore size distribution curve (inset) of TiO2 nanowire microspheres.
following calcination. This would benefit the photocatalytic activity of the microspheres. A systematic study was carried out to investigate the formation mechanisms of the TiO2 nanowire microspheres. Firstly, the role of F127 on the formation of TiO2 nanowire microspheres was studied by varying its concentration in the nanowire feed suspension prepared for spray drying. Four different concentrations of F127 solutions, namely 0, 0.01, 0.05 and 0.2 wt%, were used to prepare the nanowire suspension and the FESEM images of their products are shown in Fig. 4a–d, respectively. In the absence of F127 (0 wt%), few microspheres were formed (Fig. 4a). After increasing the F127 concentration to 0.01 wt%, some microspheres started to appear as shown in Fig. 4b. When the F127 concentration reached 0.05 wt%, few dispersed nanowires were visible and most of the nanowires have been assembled as microspheres as shown in Fig. 4c. At 0.1 wt% concentration of F127 solution, the microspherical products were more uniform in size and morphology as shown in Fig. 1a. Further increment of the F127 concentration to 0.2 wt% did not result in significant difference to the size and structure. However, when the F127 concentration was greater than 0.3 wt%, the suspension became too viscous to be dried and thus difficult to attain the required microspherical structure. These observations above concluded that the presence of F127 is critical to the formation of nanowire microsphere and the optimum concentration of F127 added is 0.1 wt%. Triblock copolymer F127 is a nonionic surfactant terminating in primary hydroxyl groups. In order to investigate the effect of the ionic type of polymer on the fabrication of the TiO2 nanowire microspheres, TiO2 microspheres were also fabricated by using 0.1 wt% anionic surfactant sodium dodecyl sulphate (SDS) and cationic surfactant cetylpyridinium chloride (CPC), respectively. The results were shown in Fig. 5. Interestingly, there is no apparent difference in the TiO2 microspheres produced using SDS or CPC. These microspheres are also similar to those fabricated using F127. This indicates that the difference in ionic types of surfactant causes no significant influence on the formation of nanowire microspheres. As shown above, the presence of surfactant is important for the formation of microspheres. The surfactant would remain in the assynthesized microsphere after spray drying. To investigate the change of surfactant during calcination, TGA was used to monitor the weight change of nanowire microspheres with temperatures increment of 30–700 8C. The initial weight ratio of F127 to nanowire in the suspension feed was 1:9. As shown in Fig. 6, the
X. Zhang et al. / Materials Research Bulletin 44 (2009) 1070–1076
1073
Fig. 4. FESEM images of the TiO2 nanowire microspheres fabricated using different concentration F127 solution. (a) 0 wt%, (b) 0.01 wt%, (c) 0.05 wt% and (d) 0.2 wt%.
Fig. 5. FESEM images of the TiO2 nanowire microspheres using different surfactant. (a) 0.1 wt% anionic surfactant SDS and (b) 0.1 wt% cationic surfactant CPC.
Fig. 6. Thermogravimetric analysis (TGA) curves for nanowire microspheres.
surfactants were mainly removed at two critical temperatures, namely, 140 and 330 8C. Upon 350 8C, all surfactants have been almost removed from the nanowire microspheres. In order to investigate the effect of varying nanowire concentration in the suspension feed on the formation of nanowire microspheres, four suspensions of different nanowire concentrations prepared using 0.1 wt% F127 were used and the FESEM images of the resulting products are shown in Fig. 7a–h. The nanowires are unable to assemble to form microspheres at low nanowire concentration, e.g. 2 g/L (Fig. 7a and b) since too less nanowire can be served for the assembly of microsphere during spray drying. As the nanowire concentration increased to 4 g/L, large amounts of microspheres were visible in the products as shown in Fig. 7c. High magnification image (Fig. 7d) clearly showed that the microspheres are hollow with an opening on the shell. The shell consisting of nanowires presents a porous morphology. At a nanowire concentration of 6 g/L, there was a further increase in quantities of the microspheres. However, the opening was absent from the shell as shown in Fig. 7e. High magnification image (Fig. 7f) shows the porous morphology of the microsphere shell
1074
X. Zhang et al. / Materials Research Bulletin 44 (2009) 1070–1076
Fig. 7. FESEM images of TiO2 nanowire microsphere fabricated using different concentration nanowire suspensions, (a and b) 2 g/L, (c and d) 4 g/L, (e and f) 6 g/L and (g and h) 10 g/L.
remained. Through the net pores between nanowires, the hollow structure is also clearly visible. At a nanowire concentration of 8 g/ L, the microspheres became close-grained as shown in Fig. 1a and b. The pore size of the microspheres also became smaller. Further increase in the concentration of nanowires to 10 g/L resulted in
formation of more close-grained microspheres as shown in Fig. 7g and h. The change in size of the microspheres is less apparent than the change in shape and structure. With the increase in nanowire concentration from 4 to 10 g/L, only a slight increase in microspherical size was recorded.
X. Zhang et al. / Materials Research Bulletin 44 (2009) 1070–1076
According to the experimental results and discussion above, the possible mechanism of the formation of TiO2 nanowire microspheres is illustrated in Fig. 8. The structure of the nanowire microsphere is critically dependent on the concentration of nanowire and surfactant presented in the suspension feed for spray drying. In the absence of surfactant, the nanowires are easily aggregated side by side because of their high specific surface energy. This may explain the absence of microspheres in the products of 0 wt% F127. In the presence of surfactant, the nanowires are dispersed evenly with ultrasonic assistance. The dissolved surfactant in water forms shells surrounding nanowires to prevent the aggregation of nanowires and well disperse them. The suspension feed of nanowires is pumped into the nozzle of the spray dryer and forced out of the orifice as a spray jet dispersion, which consists of droplets of nanowire suspension. As soon as the droplets come into contact with the air in the drying chamber, solvent evaporation takes place. Due to evaporation of water, the droplets shrink gradually and the concentrations of nanowire and surfactant in the droplets increase. With increasing concentration of surfactant, the droplets became more viscous. With the decrease in droplet diameter, the nanowires assemble towards the center of the droplet under surface tension. Nanowires then accumulate on the surface of the droplet, forming a microspherical shell. The structure of the nanowire microsphere is dependent on the nanowire concentration in the suspension feed for spray drying. If the nanowire concentration is very low (for example 2 g/L), there is insufficient nanowire content in the droplets to form the microspherical shell when the solvent is completely evaporated, therefore yielding no microspherical products. At a low concentration (for example 4 g/L), an incomplete shell will be formed, leaving an opening to the hollow interior. However, at a high nanowire concentration, a shell will form before the solvent
Fig. 8. Mechanism model of the formation of TiO2 nanowire microspheres via surfactant-mediated and spray drying process. (a) Nanowire suspension droplet without surfactant, (b) nanowire suspension droplet with low nanowire concentration, (c) nanowire suspension droplet with medium nanowire concentration and (d) nanowire suspension droplet with high nanowire concentration.
1075
evaporates. Subsequently, after formation of the shell, the droplet will continue to shrink as the solvent is being removed. The shell will be compressed further and more nanowires will accumulate on the shell, resulting in a thicker shell. Surfactant is removed from the microspheres by subsequent calcination and crystallization resulted in less elastic nanowires, hence retaining the spherical shape of the end products. To demonstrate the photoactivity in degrading contaminants from water, three kinds of nanowire microspheres, designated as NM4, NM6 and NM8, were fabricated using 4, 6 and 8 g/L nanowire suspension feeds, respectively, and were used to investigate the photocatalytic degradation of MB. The changes in MB concentration during photocatalytic degradation are shown in Fig. 9. For comparison purposes, UV irradiation without photocatalyst (photolysis) and with P25 were also carried out. In photolysis, MB reduction was less than 30% after an irradiation time of 90 min. The photocatalytic degradation of MB in TiO2 nanowire microspheres and P25 suspensions fitted the pseudo-first-order kinetic model. The apparent rate constants for the three kinds of nanowire microspheres NM4, NM6 and NM8 were 0.0639, 0.0632 and 0.0621 min 1, respectively. These were better than that for P25 (0.0556 min 1). In aqueous media, P25 TiO2 nanoparticles aggregate easily to form submicron aggregates due to their high surface energy, and this results in the reduction of contact area with UV and organic reactants. In contrast, the nanowire microspheres did not have this problem. The TOC results also showed similar results as shown in Fig. 9b. In the case of TiO2 nanowire microspheres, the nanowires were assembled by ‘‘weaved’’
Fig. 9. Changes in MB concentration and TOC during the course of photolysis and photocatalytic degradation of MB in the presence of P25 and different TiO2 nanowire microspheres. (a) MB removal and (b) TOC removal.
1076
X. Zhang et al. / Materials Research Bulletin 44 (2009) 1070–1076
For each cycle, 90 min of irradiation was applied and after the microspheres have settled by gravity, the supernatant was dumped and refilled by fresh MB solution of the same concentration. After 7 cycles, very small amounts of broken nanowire microspheres were detected. MB and TOC removal in every cycle are shown in Fig. 10. There was no significant decrease in the MB removal rates. These results indicate that the NM8 exhibited stable photocatalytic activity and mechanical strength, which are pertinent factors in cost-effective engineering applications. 4. Conclusion Hierarchically porous TiO2 nanowire microspheres with a diameter of 3–8 mm and large BET surface area of 38.2 m2/g, consisting of nanowires of 20–100 nm in diameter, were fabricated using a simple hydrothermal-spray drying-calcination process with the assistance of surfactant. The hierarchically macro/ mesoporous structure not only enhances the surface area but also allows the UV light to penetrate through the TiO2 microspheres. These synthesized microspheres show high photocatalytic activity, easily recovery by gravity, excellent stability and mechanical strength after repeated testing, indicating its high potential in engineering applications. We believe that this versatile approach can be extended to fabricate other nanowires or nanofibers into novel and efficient microspherical materials. References
Fig. 10. Cyclic runs in photocatalytic degradation of MB in the presence of TiO2 nanowire microspheres. (a) MB removal and (b) TOC removal.
nanowires, forming macropores within the shells of the microspheres. These macroporous channels could serve as light-transfer paths, which means the macroporous structure of TiO2 microsphere would allow UV light to penetrate through its shell, hence the interior of microspheres could be effectively utilized as well. This would also explain why there was no significant difference among the photocatalytic activities of NM4, NM6 and NM8, although they possessed different shell wall thickness. The TiO2 nanowire microspheres settled rapidly in the aqueous solution after air bubbling was ceased, and this resulted in efficient separation of the catalysts from the treated water. After 90 min of photocatalytic reaction, the photocatalysts were recovered for SEM characterization analysis. Micrographs showed collapse of some NM4 and NM6 nanowire microspheres, probably due to low mechanical strength of their thin shell. However, NM8 microspheres had minimum collapse and retained their spherical shapes very well. To further investigate the stability of NM8 on its photocatalytic activity and mechanical strength, NM8 were repeatedly used in photocatalytic degradation of MB for 7 cycles.
[1] M.R. Hoffmann, S.T. Martin, W. Choi, D.W. Bahnemann, Chem. Rev. 95 (1995) 69–96. [2] A. Fujishima, T.N. Rao, D.A. Tryk, J. Photochem. Photobiol., C 1 (2000) 1–21. [3] T. Kasuga, M. Hiramatsu, A. Hoson, T. Sekino, K. Niihara, Langmuir 14 (1998) 3160–3163. [4] Z. Liu, D.D. Sun, P. Guo, J.O. Leckie, Nano Lett. 7 (2007) 1081–1085. [5] X. Quan, S. Yang, X. Ruan, H. Zhao, Environ. Sci. Technol. 39 (2005) 3770–3775. [6] T.C. Long, N. Saleh, R.D. Tilton, G.V. Lowry, B. Veronesi, Environ. Sci. Technol. 40 (2006) 4346–4352. [7] S. Nagaoka, Y. Hamasaki, S.-i. Ishihara, M. Nagata, K. Iio, C. Nagasawa, H. Ihara, J. Mol. Catal. A: Chem. 177 (2002) 255–263. [8] X.Z. Li, H. Liu, L.F. Cheng, H.J. Tong, Environ. Sci. Technol. 37 (2003) 3989–3994. [9] B. Zhang, B. Chen, K. Shi, S. He, X. Liu, Z. Du, K. Yang, Appl. Catal., B 40 (2003) 253–258. [10] X. Zhang, Y. Wang, G. Li, J. Mol. Catal. A: Chem. 237 (2005) 199–205. [11] X. Zhang, G. Li, Y. Wang, Dyes Pigm. 74 (2007) 536–544. [12] G.K. Mor, O.K. Varghese, M. Paulose, K. Shankar, C.A. Grimes, Sol. Energy Mater. Sol. Cells 90 (2006) 2011–2075. [13] X. Wang, J.C. Yu, C. Ho, Y. Hou, X. Fu, Langmuir 21 (2005) 2552–2559. [14] D. Grosso, G.J.d.A.A. Soler-Illia, E.L. Crepaldi, B. Charleux, C. Sanchez, Adv. Func. Mater. 13 (2003) 37–42. [15] F. Iskandar, A.B.D. Nandiyanto, K. Myoung Yun, C.J. Hogan Jr., K. Okuyama, P. Biswas, Adv. Mater. 19 (2007) 1408–1412. [16] X. Zhang, J.H. Pan, A.J.H. Du, P.F. Lee, D.D. Sun, J.O. Leckie, Chem. Lett. 37 (2008) 424–425. [17] Z.-Y. Yuan, B.-L. Su, Colloids Surf. A 241 (2004) 173–183. [18] R. Yoshida, Y. Suzuki, S. Yoshikawa, J. Solid State Chem. 178 (2005) 2179–2185. [19] G.H. Du, Q. Chen, P.D. Han, Y. Yu, L.M. Peng, Phys. Rev. B 67 (2003) 035323. [20] H. Zhang, G.R. Li, L.P. An, T.Y. Yan, X.P. Gao, H.Y. Zhu, J. Phys. Chem. C 111 (2007) 6143–6148. [21] S. Pavasupree, Y. Suzuki, S. Yoshikawa, R. Kawahata, J. Solid State Chem. 178 (2005) 3110–3116.
Contents lists available at ScienceDirect
Materials Research Bulletin journal homepage: www.elsevier.com/locate/matresbu
Fabrication and photocatalytic activity of porous TiO2 nanowire microspheres by surfactant-mediated spray drying process Xiwang Zhang a,*, Jia Hong Pan a, Alan Jianhong Du a, Jiawei Ng a, Darren D. Sun a,*, James O. Leckie b a b
School of Civil and Environmental Engineering, Nanyang Technological University, 639798 Singapore, Singapore Department of Civil and Environmental Engineering, Stanford University, Stanford, CA 94305-4020, USA
A R T I C L E I N F O
A B S T R A C T
Article history: Received 30 June 2008 Received in revised form 18 September 2008 Accepted 24 October 2008 Available online 5 November 2008
A novel, porous TiO2 nanowire microsphere with a diameter of 3–8 mm was successfully fabricated via spray drying of TiO2 nanowire suspension with the assistance of surfactant (F127). The products were characterized by FESEM, XRD and N2 adsorption–desorption analysis and results revealed that the resulting TiO2 nanowire microspheres possessed a hierarchically macro/mesoporous structure, as well as a high BET surface area of 38.2 m2/g. Systematic studies showed that the presence of surfactant in the suspension feed for spray drying was critical in the formation of porous microspheres. The structure of the fabricated microspheres depends on the nanowire concentration in the feed. The TiO2 nanowire microspheres exhibited significant photocatalytic degradation of Methylene blue (MB) as compared to commercial TiO2 nanoparticles (P25). It was also revealed that the microspheres have excellent stability on photocatalytic activity and mechanical strength, which are both crucial factors when considering reuse of these photocatalysts. ß 2008 Elsevier Ltd. All rights reserved.
Keywords: A. Nanostructures A. Semiconductors C. Electron microscopy D. Catalytic properties
1. Introduction Due to its unique physicochemical properties, TiO2 has received considerable attention in areas of environmental purification, solar cell and gas sensor, etc. [1,2]. Recently, one-dimensional (1D) nanostructured TiO2 has gained greater popularity due to its superior performances relative to conventional bulk materials, as a result of its large surface area and nanosize effect. Recently 1D TiO2 with various morphologies (wires, fibers and tubes) have been fabricated [3–5]. However, the practical environmental applications of these 1D TiO2 materials is a huge problem. They are too small to be separated and reclaimed by conventional separation methods. From an engineering point of view, besides superior photocatalytic activity, proper photocatalysts should be easily reclaimed for repeated usages. Otherwise, it would be deemed impractical because too much TiO2 catalysts would be expended. Additionally, the unreclaimed TiO2 might result in secondary pollution [6]. Micron-sized solid spherical TiO2 photocatalysts have been synthesized by some researchers [7–11] as an ideal candidate for
* Corresponding authors. Tel.: +65 6790 6273; fax: +65 6791 0676. E-mail addresses: [email protected] (X. Zhang), [email protected] (D.D. Sun). 0025-5408/$ – see front matter ß 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.materresbull.2008.10.017
easy separation and recovery, since they can settle down easily in aqueous suspensions by gravity. Unfortunately, a significant loss in contact area between the microspherical photocatalysts and UV light limits the photocatalytic efficiency. Grime and co-workers [12] and Wang et al. [13] proved that macroporous channels in TiO2 could serve as light-transfer paths for the distribution of photon energy, which would improve the photoactivity. More recently, template-assisted methodology has been developed to fabricate TiO2 microspheres with meso/macroporous framework. Sanchez’s group has prepared multi-scale porous submicron-sized TiO2 spheres by spray drying a Ti-sol containing supremolecular template (Pluronic F127) [14]. Okuyama’s group has fabricated brookite TiO2 multi-scale porous spheres by spray drying a suspension of brookite nanoparticles and polystyrene latex (PSL) particles [15]. In our previous work [16], we found that TiO2 (B) nanowire basketry-like microspheres can be synthesized by spray drying a suspension of nanowire and polythethylene glycol (PEG). In this paper, the method was continually studied to fabricate TiO2 nanowire microsphere with controllable structure with the assistance of surfactants. The formation mechanism of the TiO2 nanowire microspheres was studied by investigating the effects of different operating parameters. Thereafter, its activity and stability were evaluated by the batch photocatalytic oxidation of a model pollutant.
X. Zhang et al. / Materials Research Bulletin 44 (2009) 1070–1076
2. Experimental 2.1. Fabrication of the TiO2 nanowire microspheres The fabrication of nanowires was commonly referred in literatures [17–19]. In a typical synthesis, TiO2 powder (Degussa, P25) was mixed with 10 M NaOH solution in a Teflon-lined autoclave container and placed in the oven at 180 8C for 2 days to undergo a hydrothermal reaction. The white pulp-like product in the autoclave was washed with 0.1 M HCl, with the assistance of ultrasound. Subsequently, pH was neutralized by repeated washing with deionized water. The powdered nanowires were added into 0.1 wt% aqueous Pluronic F127 solution to obtain 8 g/L TiO2 suspension feed for spray dryer (EYELA SD-1000). As-synthesized TiO2 nanowire microspheres were constructed by spray drying the suspension feed. Finally, they were calcined at 300–600 8C for 2 h with a heating ramp of 2 8C/min to obtain the final nanowire microspheres. 2.2. Characterization of the TiO2 nanowire microspheres A JEOL 6340 field emission scanning electron microscopy (FESEM) and a JEOL 2010 transmission electron microscopy (TEM) were used to observe the morphologies of the TiO2 nanowire microspheres. The crystal structures and the phase compositions of the samples were identified using a Bruker AXS D8 Advance Xray diffractometer with monochromated high-intensity Cu Ka irradiation (l = 1.5406 A˚) at a scanning rate of 28/min. The N2 adsorption–desorption isotherms were obtained at liquid nitrogen temperature (77 K) using a Quantachrome Autosorb1 instrument. Before the measurement, the samples were outgassed under vacuum for 5 h at 150 8C. Specific surface area was calculated by
1071
the Brunaur–Emmett–Teller (BET) method using the adsorption data at the range from P/P0 = 0.05 to 0.35 just below the capillary condensation, and the pore diameter distribution curve was derived from the adsorption branch by the BJH method. A thermalgravimeter analyzer (TGA) was used to monitor the degradation of surfactant in the sample. Measurements were taken with a heating rate of 5 8C/min from 30 to 700 8C. 2.3. Evaluation of the photocatalytic activities of TiO2 nanowire microspheres Methylene blue (MB) was used as the model chemical to investigate the photocalytic activity of the TiO2 nanowire microspheres. An Upland 3SC9 Pen-ray lamp (254 nm) was immersed into the solution as the UV source. Air was pumped into the solution to mix for the catalysts and solution, as well as to induce oxygen into the system for oxidation. The aqueous system containing MB (20 mg/L, 500 mL) and TiO2 nanowire microspheres (0.5 g/L) was mixed in the dark for 30 min to attain the adsorption equilibrium of MB with the photocatalyst, before the UV lamp was switched on. Commercial TiO2 (Degussa P25) was adopted as the reference for comparison. The characteristic absorption at l = 670 nm was chosen to monitor the concentration of MB during the photodegration process. TOC (total organic carbon) was measured using a Shimadzu TOC-5000 analyzer. 3. Results and discussion FESEM images of the as-synthesized and calcined TiO2 samples at different magnifications are shown in Fig. 1. From the low magnification image (Fig. 1a) of the as-synthesized sample, it was found that microspheres, having 3–10 mm in diameter, were produced in vast amounts. The spherical structure of the TiO2
Fig. 1. FESEM images of (a and b) as-synthesized and (c and d) calcined nanowire microspheres at 600 8C. Nanowire and F127 concentration in suspension feed: 8 g/L and 0.1 wt%.
1072
X. Zhang et al. / Materials Research Bulletin 44 (2009) 1070–1076
photocatalysts retained without apparent decrease in diameter even calcined at 600 8C (Fig. 1c). A typical FESEM image of a single as-synthesized microsphere is presented in Fig. 1b, in which it can be easily seen that TiO2 nanowires were enwrapped by the surfactant to form a microsphere. After calcination the surfactants were completely removed and a porous structure was achieved in the microspheres (Fig. 1d). The crystal structures and the phase compositions of the nanowire microspheres calcined at different temperatures were identified by wide-angle XRD as shown in Fig. 2. The diffraction pattern of the as-synthesized nanowires is similar to that as expected for protonated titanate (H2Ti2O5) with lepidocrociterelated layered structure (JCPDS 47-0124). No additional diffraction peaks of other impurities, including the anatase or rutilephased TiO2, and NaCl were detected. These results are consistent with the previous reports [20,21]. Upon calcination at 300 8C, some XRD peaks for protonated titanate became weaker or disappeared, while some new diffraction peaks of TiO2 (B), (JCPDS 35-0088) appeared. The dehydration of protonated titanate and the phase transition process were accelerated by sintering the samples at higher temperatures. Upon calcination at 500 8C, the XRD peaks of protonated titanate completely disappeared and the resulting XRD peak positions match well with the diffraction of the TiO2 (B), indicating that the protonated titanates were fully dehydrated and converted into TiO2 (B). Further increase in calcination temperatures would result in phase transformation from TiO2 (B) to anatase (JCPDS 21-1272). At 550 8C, anatase phase was found to coexist with TiO2 (B). Pure anatase phase achieved when the sample was calcined at 600 8C. The texture of TiO2 nanowire microspheres was further understood by surface area analysis and related pore-size distribution. Fig. 3 shows the nitrogen adsorption–desorption isotherm of TiO2 calcined nanowire microsphere. BET surface area was determinated to be 38.2 m2 g 1, which is very close to that of TiO2 nanowire powder samples obtained without spray drying (39.6 m2 g 1). As shown in Fig. 1b, the interlaced nanowires constitute to the walls of the microspheres and create macropores with pore sizes ranging from 50 to 150 nm. Moreover, the insert of Fig. 3 shows the pore-size distribution plot of TiO2 nanowire microspheres exhibiting a mean pore diameter of 5 nm. Thus, the multi-scale macro/meso-porosity of the resulting TiO2 nanowire microspheres is high enough to prevent a decline in surface area values after spray drying and
Fig. 2. XRD patterns of the nanowire microspheres calcined at different temperature.
Fig. 3. Nitrogen adsorption and desorption isotherms and pore size distribution curve (inset) of TiO2 nanowire microspheres.
following calcination. This would benefit the photocatalytic activity of the microspheres. A systematic study was carried out to investigate the formation mechanisms of the TiO2 nanowire microspheres. Firstly, the role of F127 on the formation of TiO2 nanowire microspheres was studied by varying its concentration in the nanowire feed suspension prepared for spray drying. Four different concentrations of F127 solutions, namely 0, 0.01, 0.05 and 0.2 wt%, were used to prepare the nanowire suspension and the FESEM images of their products are shown in Fig. 4a–d, respectively. In the absence of F127 (0 wt%), few microspheres were formed (Fig. 4a). After increasing the F127 concentration to 0.01 wt%, some microspheres started to appear as shown in Fig. 4b. When the F127 concentration reached 0.05 wt%, few dispersed nanowires were visible and most of the nanowires have been assembled as microspheres as shown in Fig. 4c. At 0.1 wt% concentration of F127 solution, the microspherical products were more uniform in size and morphology as shown in Fig. 1a. Further increment of the F127 concentration to 0.2 wt% did not result in significant difference to the size and structure. However, when the F127 concentration was greater than 0.3 wt%, the suspension became too viscous to be dried and thus difficult to attain the required microspherical structure. These observations above concluded that the presence of F127 is critical to the formation of nanowire microsphere and the optimum concentration of F127 added is 0.1 wt%. Triblock copolymer F127 is a nonionic surfactant terminating in primary hydroxyl groups. In order to investigate the effect of the ionic type of polymer on the fabrication of the TiO2 nanowire microspheres, TiO2 microspheres were also fabricated by using 0.1 wt% anionic surfactant sodium dodecyl sulphate (SDS) and cationic surfactant cetylpyridinium chloride (CPC), respectively. The results were shown in Fig. 5. Interestingly, there is no apparent difference in the TiO2 microspheres produced using SDS or CPC. These microspheres are also similar to those fabricated using F127. This indicates that the difference in ionic types of surfactant causes no significant influence on the formation of nanowire microspheres. As shown above, the presence of surfactant is important for the formation of microspheres. The surfactant would remain in the assynthesized microsphere after spray drying. To investigate the change of surfactant during calcination, TGA was used to monitor the weight change of nanowire microspheres with temperatures increment of 30–700 8C. The initial weight ratio of F127 to nanowire in the suspension feed was 1:9. As shown in Fig. 6, the
X. Zhang et al. / Materials Research Bulletin 44 (2009) 1070–1076
1073
Fig. 4. FESEM images of the TiO2 nanowire microspheres fabricated using different concentration F127 solution. (a) 0 wt%, (b) 0.01 wt%, (c) 0.05 wt% and (d) 0.2 wt%.
Fig. 5. FESEM images of the TiO2 nanowire microspheres using different surfactant. (a) 0.1 wt% anionic surfactant SDS and (b) 0.1 wt% cationic surfactant CPC.
Fig. 6. Thermogravimetric analysis (TGA) curves for nanowire microspheres.
surfactants were mainly removed at two critical temperatures, namely, 140 and 330 8C. Upon 350 8C, all surfactants have been almost removed from the nanowire microspheres. In order to investigate the effect of varying nanowire concentration in the suspension feed on the formation of nanowire microspheres, four suspensions of different nanowire concentrations prepared using 0.1 wt% F127 were used and the FESEM images of the resulting products are shown in Fig. 7a–h. The nanowires are unable to assemble to form microspheres at low nanowire concentration, e.g. 2 g/L (Fig. 7a and b) since too less nanowire can be served for the assembly of microsphere during spray drying. As the nanowire concentration increased to 4 g/L, large amounts of microspheres were visible in the products as shown in Fig. 7c. High magnification image (Fig. 7d) clearly showed that the microspheres are hollow with an opening on the shell. The shell consisting of nanowires presents a porous morphology. At a nanowire concentration of 6 g/L, there was a further increase in quantities of the microspheres. However, the opening was absent from the shell as shown in Fig. 7e. High magnification image (Fig. 7f) shows the porous morphology of the microsphere shell
1074
X. Zhang et al. / Materials Research Bulletin 44 (2009) 1070–1076
Fig. 7. FESEM images of TiO2 nanowire microsphere fabricated using different concentration nanowire suspensions, (a and b) 2 g/L, (c and d) 4 g/L, (e and f) 6 g/L and (g and h) 10 g/L.
remained. Through the net pores between nanowires, the hollow structure is also clearly visible. At a nanowire concentration of 8 g/ L, the microspheres became close-grained as shown in Fig. 1a and b. The pore size of the microspheres also became smaller. Further increase in the concentration of nanowires to 10 g/L resulted in
formation of more close-grained microspheres as shown in Fig. 7g and h. The change in size of the microspheres is less apparent than the change in shape and structure. With the increase in nanowire concentration from 4 to 10 g/L, only a slight increase in microspherical size was recorded.
X. Zhang et al. / Materials Research Bulletin 44 (2009) 1070–1076
According to the experimental results and discussion above, the possible mechanism of the formation of TiO2 nanowire microspheres is illustrated in Fig. 8. The structure of the nanowire microsphere is critically dependent on the concentration of nanowire and surfactant presented in the suspension feed for spray drying. In the absence of surfactant, the nanowires are easily aggregated side by side because of their high specific surface energy. This may explain the absence of microspheres in the products of 0 wt% F127. In the presence of surfactant, the nanowires are dispersed evenly with ultrasonic assistance. The dissolved surfactant in water forms shells surrounding nanowires to prevent the aggregation of nanowires and well disperse them. The suspension feed of nanowires is pumped into the nozzle of the spray dryer and forced out of the orifice as a spray jet dispersion, which consists of droplets of nanowire suspension. As soon as the droplets come into contact with the air in the drying chamber, solvent evaporation takes place. Due to evaporation of water, the droplets shrink gradually and the concentrations of nanowire and surfactant in the droplets increase. With increasing concentration of surfactant, the droplets became more viscous. With the decrease in droplet diameter, the nanowires assemble towards the center of the droplet under surface tension. Nanowires then accumulate on the surface of the droplet, forming a microspherical shell. The structure of the nanowire microsphere is dependent on the nanowire concentration in the suspension feed for spray drying. If the nanowire concentration is very low (for example 2 g/L), there is insufficient nanowire content in the droplets to form the microspherical shell when the solvent is completely evaporated, therefore yielding no microspherical products. At a low concentration (for example 4 g/L), an incomplete shell will be formed, leaving an opening to the hollow interior. However, at a high nanowire concentration, a shell will form before the solvent
Fig. 8. Mechanism model of the formation of TiO2 nanowire microspheres via surfactant-mediated and spray drying process. (a) Nanowire suspension droplet without surfactant, (b) nanowire suspension droplet with low nanowire concentration, (c) nanowire suspension droplet with medium nanowire concentration and (d) nanowire suspension droplet with high nanowire concentration.
1075
evaporates. Subsequently, after formation of the shell, the droplet will continue to shrink as the solvent is being removed. The shell will be compressed further and more nanowires will accumulate on the shell, resulting in a thicker shell. Surfactant is removed from the microspheres by subsequent calcination and crystallization resulted in less elastic nanowires, hence retaining the spherical shape of the end products. To demonstrate the photoactivity in degrading contaminants from water, three kinds of nanowire microspheres, designated as NM4, NM6 and NM8, were fabricated using 4, 6 and 8 g/L nanowire suspension feeds, respectively, and were used to investigate the photocatalytic degradation of MB. The changes in MB concentration during photocatalytic degradation are shown in Fig. 9. For comparison purposes, UV irradiation without photocatalyst (photolysis) and with P25 were also carried out. In photolysis, MB reduction was less than 30% after an irradiation time of 90 min. The photocatalytic degradation of MB in TiO2 nanowire microspheres and P25 suspensions fitted the pseudo-first-order kinetic model. The apparent rate constants for the three kinds of nanowire microspheres NM4, NM6 and NM8 were 0.0639, 0.0632 and 0.0621 min 1, respectively. These were better than that for P25 (0.0556 min 1). In aqueous media, P25 TiO2 nanoparticles aggregate easily to form submicron aggregates due to their high surface energy, and this results in the reduction of contact area with UV and organic reactants. In contrast, the nanowire microspheres did not have this problem. The TOC results also showed similar results as shown in Fig. 9b. In the case of TiO2 nanowire microspheres, the nanowires were assembled by ‘‘weaved’’
Fig. 9. Changes in MB concentration and TOC during the course of photolysis and photocatalytic degradation of MB in the presence of P25 and different TiO2 nanowire microspheres. (a) MB removal and (b) TOC removal.
1076
X. Zhang et al. / Materials Research Bulletin 44 (2009) 1070–1076
For each cycle, 90 min of irradiation was applied and after the microspheres have settled by gravity, the supernatant was dumped and refilled by fresh MB solution of the same concentration. After 7 cycles, very small amounts of broken nanowire microspheres were detected. MB and TOC removal in every cycle are shown in Fig. 10. There was no significant decrease in the MB removal rates. These results indicate that the NM8 exhibited stable photocatalytic activity and mechanical strength, which are pertinent factors in cost-effective engineering applications. 4. Conclusion Hierarchically porous TiO2 nanowire microspheres with a diameter of 3–8 mm and large BET surface area of 38.2 m2/g, consisting of nanowires of 20–100 nm in diameter, were fabricated using a simple hydrothermal-spray drying-calcination process with the assistance of surfactant. The hierarchically macro/ mesoporous structure not only enhances the surface area but also allows the UV light to penetrate through the TiO2 microspheres. These synthesized microspheres show high photocatalytic activity, easily recovery by gravity, excellent stability and mechanical strength after repeated testing, indicating its high potential in engineering applications. We believe that this versatile approach can be extended to fabricate other nanowires or nanofibers into novel and efficient microspherical materials. References
Fig. 10. Cyclic runs in photocatalytic degradation of MB in the presence of TiO2 nanowire microspheres. (a) MB removal and (b) TOC removal.
nanowires, forming macropores within the shells of the microspheres. These macroporous channels could serve as light-transfer paths, which means the macroporous structure of TiO2 microsphere would allow UV light to penetrate through its shell, hence the interior of microspheres could be effectively utilized as well. This would also explain why there was no significant difference among the photocatalytic activities of NM4, NM6 and NM8, although they possessed different shell wall thickness. The TiO2 nanowire microspheres settled rapidly in the aqueous solution after air bubbling was ceased, and this resulted in efficient separation of the catalysts from the treated water. After 90 min of photocatalytic reaction, the photocatalysts were recovered for SEM characterization analysis. Micrographs showed collapse of some NM4 and NM6 nanowire microspheres, probably due to low mechanical strength of their thin shell. However, NM8 microspheres had minimum collapse and retained their spherical shapes very well. To further investigate the stability of NM8 on its photocatalytic activity and mechanical strength, NM8 were repeatedly used in photocatalytic degradation of MB for 7 cycles.
[1] M.R. Hoffmann, S.T. Martin, W. Choi, D.W. Bahnemann, Chem. Rev. 95 (1995) 69–96. [2] A. Fujishima, T.N. Rao, D.A. Tryk, J. Photochem. Photobiol., C 1 (2000) 1–21. [3] T. Kasuga, M. Hiramatsu, A. Hoson, T. Sekino, K. Niihara, Langmuir 14 (1998) 3160–3163. [4] Z. Liu, D.D. Sun, P. Guo, J.O. Leckie, Nano Lett. 7 (2007) 1081–1085. [5] X. Quan, S. Yang, X. Ruan, H. Zhao, Environ. Sci. Technol. 39 (2005) 3770–3775. [6] T.C. Long, N. Saleh, R.D. Tilton, G.V. Lowry, B. Veronesi, Environ. Sci. Technol. 40 (2006) 4346–4352. [7] S. Nagaoka, Y. Hamasaki, S.-i. Ishihara, M. Nagata, K. Iio, C. Nagasawa, H. Ihara, J. Mol. Catal. A: Chem. 177 (2002) 255–263. [8] X.Z. Li, H. Liu, L.F. Cheng, H.J. Tong, Environ. Sci. Technol. 37 (2003) 3989–3994. [9] B. Zhang, B. Chen, K. Shi, S. He, X. Liu, Z. Du, K. Yang, Appl. Catal., B 40 (2003) 253–258. [10] X. Zhang, Y. Wang, G. Li, J. Mol. Catal. A: Chem. 237 (2005) 199–205. [11] X. Zhang, G. Li, Y. Wang, Dyes Pigm. 74 (2007) 536–544. [12] G.K. Mor, O.K. Varghese, M. Paulose, K. Shankar, C.A. Grimes, Sol. Energy Mater. Sol. Cells 90 (2006) 2011–2075. [13] X. Wang, J.C. Yu, C. Ho, Y. Hou, X. Fu, Langmuir 21 (2005) 2552–2559. [14] D. Grosso, G.J.d.A.A. Soler-Illia, E.L. Crepaldi, B. Charleux, C. Sanchez, Adv. Func. Mater. 13 (2003) 37–42. [15] F. Iskandar, A.B.D. Nandiyanto, K. Myoung Yun, C.J. Hogan Jr., K. Okuyama, P. Biswas, Adv. Mater. 19 (2007) 1408–1412. [16] X. Zhang, J.H. Pan, A.J.H. Du, P.F. Lee, D.D. Sun, J.O. Leckie, Chem. Lett. 37 (2008) 424–425. [17] Z.-Y. Yuan, B.-L. Su, Colloids Surf. A 241 (2004) 173–183. [18] R. Yoshida, Y. Suzuki, S. Yoshikawa, J. Solid State Chem. 178 (2005) 2179–2185. [19] G.H. Du, Q. Chen, P.D. Han, Y. Yu, L.M. Peng, Phys. Rev. B 67 (2003) 035323. [20] H. Zhang, G.R. Li, L.P. An, T.Y. Yan, X.P. Gao, H.Y. Zhu, J. Phys. Chem. C 111 (2007) 6143–6148. [21] S. Pavasupree, Y. Suzuki, S. Yoshikawa, R. Kawahata, J. Solid State Chem. 178 (2005) 3110–3116.