Langmuir 2007, 23, 11421-11424
11421
One-Step Synthesis of Highly Dispersed Gold Nanocrystals on Silica Spheres Nopphawan Phonthammachai* and Timothy J. White School of Materials Science and Engineering, Nanyang Technological UniVersity, Singapore 639798 ReceiVed July 24, 2007. In Final Form: September 10, 2007 Highly dispersed gold nanocrystals decorating silica spheres were prepared from HAuCl4 and NaOH via a depositionprecipitation (DP) process, in which the isoelectric point (IEP) of the substrate was adjusted during sphere synthesis by interaction of the surface with ammonia molecules. Through the systematic variation of pH (4-8), reaction temperature (65-96 °C), and time (10-30 min), a superior product with small (2-5 nm), homogeneously distributed gold crystals was obtained at pH 7 and a reaction temperature of 96 °C. These materials will offer enhanced performance as catalysts and contrast enhancers in biomedical imaging.
Introduction Ceramic spheres coated with nanocrystalline metals find application in catalysis and advanced bioimaging techniques.1-3 The most commonly studied core material is amorphous silica,4 although iron oxide5 is appropriate when magnetic separation is required. The metal overlayer is usually gold,6 but silver,7 palladium,8 and cobalt9 decoration have also been reported. Dielectric cores10-16 coated by a continuous shell of gold nanocrystals display an optical resonance that when tuned to the near-infared (NIR) enhances molecular contrast for reflectance confocal microscopy (RCM) and optical coherence microscopy (OCT)17-19 of biomaterials. Alternatively, if the overlayer is composed of dispersed gold crystals 500 °C) is required when catalysis is the end use.23 Alternatively, the deposition-precipitation (DP) method can reliably synthesize oxide-supported gold24,25 where the isoelectric point (IEP) of the support material is the key parameter. Moreau et al.26 found that the deposition rate of agglomerated gold crystals on titania accelerated with the pH < IEP as the positive surface charge attracted gold-bearing anions (AuCl4-). However, with pH > IEP the titania surface carries a net negative charge, and electrostatic repulsion of the anions yield highly dispersed gold crystals with the coverage controlled by several factors including the concentration of HAuCl4; volume to mass ratio of HAuCl4 to support; base type (NaOH, NH4OH, urea); reaction time and temperature; filtration, washing, and drying steps; and calcination temperature.27 Selection of the appropriate alkaline solution is especially critical for controlling the yield, dispersion, and size of the gold crystals. For example, Zanella et al.28 found that, when DP on titania was conducted with NaOH, the gold was partially reduced to metal after drying at 100 °C, but a comparable experiment with urea delayed metallization to temperatures >150 °C. Although DP can successfully load gold nanocrystals on metal oxides, their dispersion over silica surfaces has been problematic and the sizes often nonuniform, because the isoelectric point (IEP) of silica (pH ∼2) favors the accumulation of negative surface charge that hinders the deposition of anionic species. To overcome this, Zanella et al.21 prepared gold-loaded silica by using the cationic complex [Au(en)2]3+ to improve gold dispersion, in a difficult process that failed to completely coat the silica kernel. This paper describes a simple DP method where adjustment of surface charge by ammonia adsorption allowed the decoration of silica spheres with highly dispersed gold nanocrystals of uniform size. Experimental Section Silica Sphere Synthesis. The reagents were tetraethyl orthosilicate (TEOS, 99%, Merck), ammonia (NH3, 27%, Merck), ammonium hydroxide (NH4OH, 14.2 M, ISPL), and ethanol (EtOH, 99.9%, (21) Zanella, R.; Sandoval, A.; Santiago, P.; Basiuk, V. A.; Saniger, J. M. J. Phys. Chem. B 2006, 110, 8559. (22) Pham, T.; Jackson, J. B.; Halas, N. J.; Lee, T. R. Langmuir 2002, 18, 4915. (23) Chi, Y.-S.; Lin, H.-P.; Mou, C.-Y. Appl. Catal., A 2005, 284, 199. (24) Haruta, M.; Kobayashi, T.; Sano, H.; Yamada, N. Chem. Lett. 1987, 16, 405. (25) Haruta, M. Gold Bull. 2004, 37, 27. (26) Moreau, F.; Bond, G. C.; Taylor, A. O. J. Catal. 2005, 231, 105. (27) Dobrosz, I.; Jitratova, K.; Pitchon, V.; Rynkowski, J. M. J. Mol. Catal. A: Chem. 2005, 234, 187. (28) Zanella, R.; Delannoy, L.; Louis, C. Appl. Catal., A 2005, 291, 62.
10.1021/la702230h CCC: $37.00 © 2007 American Chemical Society Published on Web 10/04/2007
11422 Langmuir, Vol. 23, No. 23, 2007 Merck). Following the Sto¨ber method,29 TEOS/EtOH/H2O/NH3/ NH4OH (1.5:23.0:0.8:6.2:1.0 by volume) were stirred vigorously for 2 h to achieve a white turbid suspension. The spheres were separated from the mother liquor by centrifugation (8000 rpm/10 min), then washed repeatedly with deionized water and once with ethanol. The monosized spheres were redispersed in ethanol in a weight ratio of 1:4 (silica/EtOH), calcined at 700 °C for 2 h with ramp rate of 1 °C/min, and redispersed in EtOH before use. Gold Nanocrystal Loaded Silica Spheres. Reaction temperature and time and pH were varied systematically to optimize the dispersion of the crystals gold over the silica spheres. The pH (4 < pH < 8) of the gold solution was adjusted by addition of 0.1 M NaOH to 20 mL of a 6.35 mM HAuCl4 solution, and then 1 mL of the uncalcined silica sphere suspension was dispersed with vigorous stirring and warmed (65-96 °C) for specific reaction times (5-30 min). The gold-seeded particles were collected using a centrifuge (1200 rpm) and washed five times with deionized water. The particles were redispersed ultrasonically in 40 mL of deionized water prior to use. The same procedure was applied to the calcined silica spheres at pH 7 using a reaction temperature of 96 °C for 15 min. Materials Characterization. The crystallographic parameters and phase content of the silica-gold composites were determined from powder X-ray diffraction (XRD) patterns obtained with a Shimadzu diffractometer operating at 4 kW and using Cu KR radiation. Materials were prepared for diffraction by packing powder into a glass holder (2 × 2 cm) with diffracted intensity collected by step scanning over the 2θ range 10-90° in increments of 0.02° and a dwell time of 0.60 s. Rietveld refinement as implement in TOPAS 2.1 was used for quantitative phase analysis with the starting model being gold (Fm3hm).30 For each analysis, a background polynomial, scale factor, cell parameters, and zero point correction were refined to determine phase content, check for mass balance, identify phase transformations, and estimate crystal size. Transmission electron microscopy (TEM) operating in high-resolution mode was conducted using a JEOL 2100F instrument at a voltage of 200 kV. The valence band of gold nanocrystals and the ammonia adsorption were characterized using X-ray photoemission spectroscopy (XPS, PHI5600). The atomic levels of Au4f and N1s were measured to detect the energy of the chemical shifts.
Letters Table 1. Gold Nanocrystal Characteristics as a Function of pH, Reaction Temperature, and Reaction Time for Metal Decorating Amorphous Silica Spheres
sample
pHia
pHab
pHfc
reaction temp (°C)
reaction time (min)
average Au crystal sized (nm)
USiAu1 4 5.1 5.2 65 30 15 USiAu2 6.1 8.2 8.2 65 30 16 USiAu3 7.2 9.4 8.9 65 5 2 USiAu4 7.2 9.4 8.9 65 15 4 USiAu5 7.2 9.4 8.9 65 30 8 USiAu6 7.2 9.4 8.9 85 15 5 USiAu7 7.2 9.4 8.9 96 15 5 USiAu8 8 9.3 9.1 65 30 10 a Initial pH of gold solution. b pH after additional of silica sphere. c pH at the end of reaction. d Determined by TEM.
Figure 1. XPS confirmation of physisorbed ammonia on an airdried amorphous silica surface. Scheme 1. Proposed Mechanism for the Deposition of Gold Anions on Ammonia Adsorbed Silica Surfaces
Results and Discussion Effect of pH. Optimization of the pH of the HAuCl4 solution is the key factor controlling the dispersion of gold nanoparticles on metal oxides, as this regulates the hydrolysis of AuCl4- ions and the adsorption of chloride on the support. The stability of [Au(OH)4]- relative to AuCl4- is enhanced at higher pH, while adsorption of chloride behaves in a contrary fashion. Earlier studies have demonstrated that [AuCl4]- is dominant at pH
