16850
2009, 113, 16850–16854 Published on Web 09/04/2009
A pH-Induced Size Controlled Deposition of Colloidal Ag Nanoparticles on Alumina Support for Catalytic Application Kohsuke Mori, Akihito Kumami, Masanori Tomonari, and Hiromi Yamashita* DiVision of Materials and Manufacturing Science, Graduate School of Engineering, Osaka UniVersity, 2-1 Yamada-oka, Suita, Osaka 565-0871, Japan ReceiVed: July 30, 2009; ReVised Manuscript ReceiVed: August 27, 2009
A rational synthetic method of supported metal catalysts has been proposed. Highly dispersed Ag nanoparticles (NPs) with a mean diameter of ca. 10 nm and a narrow size distribution have been prepared in the presence of 3-mercaptopropionic acid (3-MPA) as a protective reagent. The zeta electric potential of the Ag NPs was negatively charged in the region of pH higher than 5 due to the presence of dissociated carboxylate ion (-COO-), which led to the electric repulsion between Ag NPs. On the other hand, the electric charge gradually increased in the region of pH less than 5 owing to the formation of a protonated carboxyl group (-COOH), which induced the hydrogen bonding between Ag NPs. Such pH-triggered assembly dispersion control enables a strong protocol to deposit Ag NPs with different diameters on the positively charged Al2O3 support by electrostatic attraction. The obtained Ag/Al2O3 materials were characterized by means of UV-vis spectra and transmission electron microscopy (TEM), and the catalytic activities were evaluated in the reduction of 4-nitrophenol in water. The reaction rate increased with increasing pH of the prepared colloidal solution, which is correlated with the decreasing size of the Ag NPs. The reduction of 4-nitrophenol is considered as a structure-insensitive reaction, and all surface atoms of the Ag NPs act catalytically as the same active species in this size range by performing calculations on the Ag crystallites. Introduction There is a great deal of interest in the utilization of metal nanoparticles (NPs) in the field of catalysis, nanoelectronics, and material science owing to their size-dependent properties.1 Their existence on the borderline molecular states with discrete quantum energy levels provides the opportunities to bridge the gap between mononuclear metal complex and heterogeneous bulk catalysts.2 Large surface area-to-volume ratio contributes to effective utilization of expensive metals. The variation in primary structures, such as size, composition, and morphology allows precise control of catalytic activities.3 Protective ligands can also impart other desired characteristics, such as selective solubility, resistance to aggregation, and derivatization with functional groups. Although the NP-based catalysts are key components of catalytic activities, the application in liquid suspensions is impractical due to the difficulties in the product separation and catalysts recycling. In order to overcome the above drawbacks, the seeking of new methodologies for the synthesis of supported metal catalysts is of fundamental industrial and academic interest.4 Parameters to control in supported metal catalysts include particle composition, size, and shape as well as characteristics of supports.5 This multidimensional problem is complex; thus, the design and synthetic control of promising catalysts is a serious technological challenge. The most convenient methods for attaining the immobilization of metal NPs on high surface area support materials are the incipient wetness technique and the ion* To whom correspondence should be addressed. E-mail: yamashita@ mat.eng.osaka-u.ac.jp. Phone and fax: +81-6-6879-7457.
10.1021/jp907277g CCC: $40.75
exchange method, in which the support material is impregnated with metal precursors in the solution phase, followed by activation under a reducing atmosphere.6 Owing to its inherent simplicity, this is successful for the large scale production of catalysts but, unfortunately, does not provide satisfactory control of particle size. There have been some reports on the grafting of preformed colloidal metal NPs by use of the Coulombic attractive force between charged colloidal particles and solid surfaces.7 The dispersions of these catalysts are preferable to the conventional supported metal catalyst in the homogeneity of their particle sizes, the reproducibility in the preparation, and activity and selectivity as catalysts. However, electrical charges are neutralized at a pH near the isoelectric point of the particles, and electrostatic interactions are drastically suppressed. In this study, we synthesized Ag NPs with a mean diameter of ca. 10 nm stabilized by 3-mercaptopropionic acid (3-MPA) as a colloidal precursor solution. Precise control of the size of the colloidal Ag NPs can be achieved by the pH-triggered reversible assembly dispersion properties. The negatively charged Ag NPs with different states of aggregation have been deposited on the positively charged Al2O3 support by electrostatic attraction while keeping their inherent aggregated form (Figure 1). These Ag/Al2O3 materials with different mean particle sizes are active for the reduction of 4-nitrophenol in water. The size effect of the Ag NPs on the catalytic activities was also investigated. This study demonstrates a new strategy in catalyst design to fabricate well-controlled catalytically active sites based on solution phase colloidal NP chemistry. 2009 American Chemical Society
Letters
J. Phys. Chem. C, Vol. 113, No. 39, 2009 16851
Figure 1. Schematics of the assembly dispersion mechanism of the colloidal Ag NPs with 3-MPA and their deposition onto Al2O3 support.
Experimental Section The Synthesis of Colloidal Ag NPs Stabilized with 3-MPA. Ag NPs were prepared using standard airless techniques on a Schlenk line. AgNO3 (0.1 mol), 14 N aq. NH4OH (26.0 mL), and 3-MPA (8 × 10-3 mol) were dissolved in deionized water (15 mL) (solution A). NaBH4 (0.02 mol) and 14 N aq. NH4OH (2.0 mL) were dissolved in deionized water (15 mL) (solution B). The above solutions A and B were slowly added to deionized water (300 mL) over 30 min at room temperature. After 30 min, the mixture was subjected to ultrafiltration, washed with water, and stored in water. The Synthesis of Ag NP/Al2O3. A colloidal Ag aqueous solution (Ag: 50 wt %) was mixed with water (50 mL), and the pH of the solution was adjusted with 0.1 N HCl. The obtained solution was quickly mixed with Al2O3 (1.0 g, JRC-ALO-8 supplied from the Japan Catalysis Society), and the slurry was stirred for 1 h at room temperature. The precipitate was separated by filtration, thoroughly washed with deionized water, and dried under a vacuum. The obtained sample was calcined at 823 K for 5 h and treated by H2 (20 mL · min-1) at 473 K for 1 h. The amount of Ag deposition was determined to be 0.6 wt % by ICP analysis. For example, the product prepared at pH 3.7 was named as Ag/Al2O3 (3.7). Characterization. X-ray diffraction patterns were recorded using a Rigaku Mini-flex using Cu KR radiation of wavelength 1.5418 Å. Elemental analysis was performed with EDX-720 (Shimadzu). The UV-vis spectra were recorded with a Shimadzu UV-2200A photospectrometer. The zeta potential measurement was carried out with ELSZ-2 (Otsuka electronics). HCl and NH4OH were used to adjust the solution pH. TEM micrographs were obtained with a Hitachi Hf-2000 FE-TEM equipped with a Kevex energy-dispersive X-ray detector operated at 200 kV. Ag K-edge XAFS spectra were recorded at room temperature in fluorescence mode at the beamline 01B1 station with an attached Si(311) monochromator at SPring-8, JASRI, Harima, Japan (prop. no. 2008A1366, 2008A1457).
A Typical Example for the Reduction of 4-Nitrophenol. Into a reaction vessel were placed Ag/Al2O3 (0.02 g) and 46.4 mM aqueous 4-nitrophenol solution (4 mL), and Ar gas was purged through the solution to remove the dissolved O2. After the addition of 13.9 mM aqueous NaBH4 solution (20 mL), the resulting mixture was stirred at 300 K. Part of the mixture was taken out after every 10 min and centrifuged for the determination with UV photospectrometer. The peak at 400 nm corresponding to the 4-nitrophenol was monitored to determine the conversion of 4-nitrophenol. Results and Discussion The synthesis of aqueous Ag colloidal solution at a very high density of Ag metal with superior stability could be realized with 3-MPA as a protective agent, which displayed the redrust color characteristic of the absorption due to the plasmon oscillation of Ag NPs.8 It is well-known that the thiol groups of 3-MPA readily form very stable metal-sulfur bonds with the surface atoms under mild conditions.9 In the XRD measurement, the only peaks due to the crystalline of Ag metal were observed. The formation of Ag metal as a single phase also could be confirmed by the X-ray absorption fine structure (XAFS) measurement at the Ag K-edge. The transmission electron microscopy (TEM) image is shown in Figure 2a. The Ag metal NPs in the spherical form with the well-controlled particle size can be observed. The distribution of the particle sizes of Ag metal is very narrow, and the average particle size was determined to be about 10.4 nm. The effect of the pH of colloidal solution on the dispersed Ag NPs has been studied by the addition of the HCl and NH4OH aqueous solutions. The stable colloidal state having red-rust color was kept while the pH was adjusted in the region of the alkaline side. On the other hand, the assembly of Ag NPs occurred accompanied with the color change from red purple to black in the region of pH less than 4. The assembly of Ag NPs became serious at pH values less than 2, and the aggregated Ag NPs grew to be separated from the solution completely.
16852
J. Phys. Chem. C, Vol. 113, No. 39, 2009
Letters
Figure 2. TEM and HR-TEM images and size distribution diagrams of (a) colloidal Ag NP with 3-MPA, (b) Ag/Al2O3 (3.7), and (c) Ag/Al2O3 (2.9).
Interestingly, the Ag NPs were redispersed to make a colloidal state with the original red-rust color by the addition of NH4OH aqueous solution. The assembly dispersion reversibility of the Ag NPs was also confirmed by the shift of the absorption due to the plasmon oscillation in the UV-vis spectra (Figure S1 of the Supporting Information).10 The above sequence was investigated by the zeta electric potential analysis. As shown in Figure 3, the surface of the Ag NPs in the solution was negatively charged in the region of pH higher than 5, but the electric charge increased in the region of pH
less than 5 and became zero at around pH 2. 3-MPA readily binds to the surface of Ag NPs via a thiolate linkage as a protective agent. In the solution at higher pH values, the proton of the carboxyl group at the opposite side of the thiol group was electrolytically dissociated to form carboxylate anion (-COO-) and a negative electric potential was formed on the surface of Ag NPs. Such Ag NPs can demonstrate the high dispersion state without the aggregation because of their electrostatic repulsion between Ag NPs. At the lower pH, the Ag NP surface becomes electrostatically neutral to lose the driving force of dispersion and the hydrophobic
Letters
J. Phys. Chem. C, Vol. 113, No. 39, 2009 16853
Figure 5. Time profile for the reduction of 4-nitrophenol. Figure 3. pH dependence of the zeta potentials for Al2O3 and Ag NPs with 3-MPA.
Figure 4. UV-vis absorption spectra of the Ag/Al2O3 prepared under various pH conditions.
interactions between Ag NPs predominate, because the proton of the weak acid carboxyl group of 3-MPA cannot be ionized. It can be envisioned that such pH-triggered assembly dispersion properties of the Ag NPs provide a possibility to control the particle size of the supported metal catalysts. We selected Al2O3 as a catalyst support. The pH dependence of the zeta potential for Al2O3 is also shown in Figure 3. The surface charge of Al2O3 was positive at low pH and negative at higher pH; the isoelectric point was between pH 8 and 10. These results suggest that negatively charged Ag NPs are easily deposited on the positively charged Al2O3 support by electrostatic attraction at pH values less than 5. With this in mind, several colloidal Ag NP aqueous solutions whose pH was adjusted with HCl (pH 2.9, 3.2, 3.4, and 3.7) were quickly mixed with a powdered Al2O3 support. As expected, Ag NPs were smoothly deposited on the Al2O3 surface by electrostatic attraction while keeping their dispersion state. After deposition of Ag NPs, the samples were calcined at 823 K to remove the surface attached 3-MPA and further treated with H2 at 473 K. The amount of Ag deposition was determined to be 0.6 wt % for all samples by ICP analysis. The UV-vis diffuse reflectance spectra of these samples are shown in Figure 4. No absorption is observed in the Al2O3 without Ag NPs. The characteristic plasmon absorption peak of Ag NPs was observed at around 450 nm. To confirm the pH-induced size control of the deposited Ag NPs, TEM measurement was carried out. The selected TEM images are shown in Figure 2, respectively. As mentioned before, the mean diameter of the colloidal Ag NPs with 3-MPA was 10.4 nm with a narrow size distribution (Figure 2a). The Ag/Al2O3(3.7) showed smaller Ag NPs in size, where the
average diameter was determined to be 12.1 nm (Figure 2b). The average diameter gradually increased with decreasing pH value in the preparation sequence; the average diameters at pH 3.4 and 3.2 were estimated to be 15.5 and 19.2 nm, respectively (Figure S2 of the Supporting Information).10 Finally, the average diameters at pH 2.9 were reached up to 26.3 nm (Figure 2c). A linear relationship was observed between the pH value in the preparation sequence and the average diameter of Ag NPs (Figure S3 of the Supporting Information).10 It can be concluded that the present preparation method based on the precise pHinduced control of the assembly dispersion state of the colloidal NPs enables a strong protocol to create supported metal catalysts with different diameters. High-resolution TEM (HR-TEM) images did not show a significant grain boundary, revealing that the Ag NPs deposited on Al2O3 were not the assemblages of the original small NPs but the single crystalline. The average crystalline size calculated by applying Scherrer’s equation was also consistent with the results of TEM analysis, verifying the single crystalline nature of the Ag NPs. Next, we evaluated the potential catalytic activities of the Ag/Al2O3 in the reduction of 4-nitrophenol to 4-aminophenol in the presence of NaBH4. No reaction was observed either in the absence of Ag catalyst or NaBH4.11 The Ag/Al2O3 (3.7) with the smallest Ag NPs exhibited higher activity compared to other Ag/Al2O3 samples (Figure 5). Because the higher dispersion of Ag metal NPs is preferable for the catalytic reactions, the results on the reduction of 4-nitrophenol also support the fact that the present preparation method is useful to control nanosized Ag metals based on the simple pH-induced assembly dispersion properties. Furthermore, the structure sensitivity of the Ag NP catalysts for the reduction of 4-nitrophenol has been investigated by relating activity data with Ag NP size.12 The particle surface is composed of different types of sites classified as high-coordination-number terrace atoms and low-coordination-number defect atoms (edge and corner). The relative proportion of these sites with respect to the total number of surface atoms is well-known to vary with the metal particle sizes. In an effort to examine the correlation between the amount of potentially active Ag sites and the catalytic activity for the reduction of 4-nitrophenol, four Ag/Al2O3 catalysts with different mean diameters were employed. Assuming that the present Ag NPs are cuboctahedral in shape with a cubic close packed structure in this size range, the model of full-shell NPs is adopted.13 The total number of the Ag atoms of NP (NT) can be calculated from eq 1 as a function of particle edge length m.14
NT ) (2m - 1)(5m2 - 5m + 3)/3
(1)
16854
J. Phys. Chem. C, Vol. 113, No. 39, 2009
Letters
TABLE 1: Geometric Parameters and TOF on Various Ag/Al2O3
average diameter/nm number of Ag atoms NT number of surface Ag atom NS number of low-coordination site NLS TOF based on NS/s-1 TOF based on NLS/s-1
Ag/Al2O3 (3.7)
Ag/Al2O3 (3.4)
Ag/Al2O3 (3.2)
Ag/Al2O3 (2.9)
12.1 55301 6252
15.5 114465 10242
19.2 221481 16002
26.3 569911 30252
588 1.6 17.3
756 1.8 24.1
948 2.0 33.8
1308 1.8 42.5
(2)
Moreover, the low-coordination-number edge and corner atoms (NLS), which are considered main active species in the “structuresensitive reaction”, are obtained according to eq 3.
NLS ) 24(m - 2) + 12
Supporting Information Available: UV-vis spectra of Ag colloidal solution with 3-MPA under different pH conditions and the TEM images of the Ag/Al2O3. This material is available free of charge via the Internet at http://pubs.acs.org. References and Notes
The numbers of surface atoms (NS) are also given by eq 2
NS ) 10m2 - 20m + 12
Acknowledgment. This work is supported by a Grant-inAid for Scientific Research from the Ministry of Education, Culture, Sports, Science, and Technology, Japan. The authors appreciate Dr. Eiji Taguchi and Prof. Hirotaro Mori at the Research Center for Ultra-High Voltage Electron Microscopy, Osaka University, for assistance with TEM measurements.
(3)
Table 1 summarizes the calculated geometric parameters for three different Ag NP catalysts, respectively. With four prepared Ag/Al2O3 having different mean diameters of Ag NPs, the TOF values were determined on the basis of the NS and NLS by measuring the conversion of 4-nitrophenol at ca. 10%, as listed in Table 1. The TOF values based on NLS depend on the particle size. Upon consideration of the surface Ag atoms (NS) as catalytically active sites, the normalized TOF values were found to be almost independent of the Ag NP sizes. It can be said that all surface Ag atoms are responsible for the catalysis in this size range. Therefore, the reduction of 4-nitrophenol is considered to be “structure-insensitive” and the smaller Ag NPs are effective to attain high catalytic activity. In summary, Ag NPs stabilized with 3-MPA were negatively charged in the region of pH higher than 5 due to the electric repulsion between Ag NPs, while electric charge gradually approached to neutral in the region of pH less than 5 owing to the hydrophobic interaction. On the basis of such pH-induced assembly dispersion properties, the deposition of Ag NPs on Al2O3 support was successfully attained by electrostatic attraction while keeping their inherent dispersion state. The Ag/Al2O3 acted as efficient catalysts for the reduction of 4-nitrophenol. No dependence between the normalized TOF and the surface Ag atoms suggests that the reaction may occur at all surface Ag atoms identically. The present strategy provides great flexibility in the selection of the kind of metals as well as primary NP sizes. Introduction of these features into the catalyst design enables achievement of desired supported metal catalysts for the target catalytic reactions.
(1) (a) Kamat, P. V.; Meisel, D. Curr. Opin. Colloid Interface Sci. 2002, 7, 282. (b) Wilcoxon, J. P.; Abrams, B. L. Chem. Soc. ReV. 2006, 35, 1162. (c) MacKenzie, J. D.; Bescher, E. P. Acc. Chem. Res. 2007, 40, 810. (d) Kwon, S. G.; Hyeon, T. Acc. Chem. Res. 2008, 41, 1696. (2) (a) El-Sayed, M. A. Acc. Chem. Res. 2001, 34, 257. (b) Astruc, D.; Lu, F.; Aranzaes, J. R. Angew. Chem., Int. Ed. 2005, 44, 7852. (c) Mori, K.; Kondo, Y.; Morimoto, S.; Yamashita, H. J. Phys. Chem. C 2008, 112, 2593. (d) Mori, K.; Sugihara, K.; Kondo, Y.; Takeuchi, T.; Morimoto, S.; Yamashita, H. J. Phys. Chem. C 2008, 112, 16478. (3) Laslie-Pelecky, D. L.; Rieke, R. D. Chem. Mater. 1996, 8, 1770. (4) (a) Ertl, G.; Knoezinger, H.; Weitkamp, J., Eds. Catalytic process. Part B. Handbook of heterogeneous catalysis; Wiley-VCH: New York, 1997. (b) Geus, J. W.; van Ween, J. A. R. Catalysis. An Integrated Approach to Homogeneous and Heterogeneous and Industrial Catalysis; Moulijin, J. A., van Leeuwen, P. W. N. M., van Santen, R. A., Eds.; Elsevier: Amsterdam, The Netherlands, 1993. (c) Kongkanand, A.; Vinodgopal, K.; Kuwabata, S.; Kamat, P. V. J. Phys. Chem. B 2006, 110, 16185. (d) Liu, Y.; Tsunoyama, H; Akita, T.; Tsukuda, T. J. Phys. Chem. C 2009, 113, 13457. (5) (a) Grunes, J.; Zhu, J.; Somorjai, G. A. Chem. Commun. 2003, 18, 2257. (b) Mori, K.; Hara, T.; Mizugaki, T.; Ebitani, K.; Kaneda, K. J. Am. Chem. Soc. 2004, 126, 10657. (c) Tsunoyama, H.; Ichikuni, N.; Sakurai, H.; Tsukuda, T. J. Am. Chem. Soc. 2009, 131, 7086. (6) (a) Hayek, K.; Kramer, R.; Paa´l, Z. Appl. Catal., A 1997, 162, 1. (b) Yang, C.-M.; Liu, P.-H.; Chiu, C.-Y.; Chao, K.-J. Chem. Mater. 2003, 15, 275. (7) (a) Ohtaki, M.; Toshima, N.; Komiyama, M.; Hirai, H. Bull. Chem. Soc. Jpn. 1990, 63, 1433. (b) Yu, K. M. K.; Yeung, C. M. Y.; Thompsett, D.; Tsang, S. C. J. Phys. Chem. B 2003, 107, 4515. (c) Rioux, R. M.; Song, H.; Hoefelmeyer, J. D.; Yang, P.; Somorjai, G. A. J. Phys. Chem. B 2005, 109, 2192. (d) Rinaldi, R; Porcari, A. M.; Rocha, T. C. R.; Cassinelli, W. H.; Ribeiro, R. U.; Bueno, J. M. C.; Zanchet, D. J. Mol. Catal. A: Chem. 2009, 301, 11. (8) (a) He, S.; Yao, J.; Jiang, P.; Shi, D.; Zhang, H.; Xie, S.; Pang, S.; Gao, H. Langmuir 2001, 17, 1571. (b) Sun, X.; Li, Y. Langmuir 2005, 21, 6019. (c) Au, J.; Tang, B.; Ning, X; Zhou, J.; Xu, S.; Zhao, B.; Xu, W; Corredor, C.; Lombardi, J. R. J. Phys. Chem. C 2007, 111, 18055. (9) Vasiliev, A. N.; Gulliver, E. A.; Khinast, J. G.; Riman, R. E. Surf. Coat. Technol. 2009, 203, 2841. (10) See the Supporting Information. (11) Liu, P.; Zhao, M. Appl. Surf. Sci. 2009, 255, 3989. (12) Van Hardeveld, R.; Hartog, F. Surf. Sci. 1969, 15, 189. (13) It should be noted that, although the particles have a size distribution, the mean diameters of particles were employed for the calculations of the geometric parameters such as NT, NS, and NLS. The similar approximation has been carried out in the studies of the Pd particle size effect on coupling reactions. See: (a) Augustine, R. L.; O’Leary, S. T. J. Mol. Catal. A: Chem. 1995, 95, 277. (b) Le Bars, J.; Specht, U.; Bradley, J. S.; Blackmond, D. G. Langmuir 1999, 15, 7621. (c) Li, Y.; Boone, E.; El-Sayed, M. A. Langmuir 2002, 18, 4921. (14) Benfield, R. E. J. Chem. Soc., Faraday Trans. 1992, 88, 1107.
JP907277G
