5924
Langmuir 2004, 20, 5924-5932
Evaluating Formation and Growth Mechanisms of Silica Particles Using Fluorescence Anisotropy Decay Analysis Dina Tleugabulova,† Andy M. Duft,‡ Zheng Zhang,† Yang Chen,† Michael A. Brook,† and John D. Brennan*,† Department of Chemistry, McMaster University, Hamilton, Ontario L8S 4M1, Canada, and Brockhouse Institute for Materials Research, McMaster University, Hamilton, Ontario L8S 4M1, Canada Received February 20, 2004. In Final Form: April 21, 2004 At present, there is no direct experimental evidence that primary silica particles, which exist only transiently for a few seconds during the Sto¨ber silica synthesis, can be stable in aqueous solutions. In the present work, we show that primary silica particles are formed spontaneously after the dissolution of diglycerylsilane (DGS) in aqueous solutions and remain stable for prolonged periods of time. By using time-resolved fluorescence anisotropy (TRFA), we demonstrate that this unique property of DGS is ascribed to the slow kinetics of silica particle growth in diluted sols at pH ∼ 9.0. The anisotropy decay of the cationic dye rhodamine 6G (R6G), which strongly adsorbs to silica oligomers and nanoparticles in DGS sols, could be fit to three components: a fast (picosecond) scale component associated with free R6G, a slower (nanosecond) rotational component associated with R6G bound to primary silica particles, and a residual (nondecaying) anisotropy component associated with R6G that was bound to secondary or larger particles that were unable to rotate on the time scale of the R6G emission lifetime (4 ns). The data show that, under conditions where fast hydrolysis is obtained, the initial size of the nuclei depends on the silica concentration, with larger nuclei being present in more concentrated sols, while the rate of growth of primary particles depends on both silica concentration and solution pH. At low silica concentrations and high pHs, it was possible to observe the growth of stable, nonaggregating primary silica particles by a mechanism involving rapid nucleation followed by monomer addition. The presence of stable primary particles was confirmed by atomic force microscopy (AFM) imaging. At higher silica concentrations and lower pHs, there was an increase in the initial size of the nuclei formed, which subsequently grew to a larger radius (>4.5 nm) or aggregated with time, and in such cases, nucleation and aggregation occurred simultaneously in the early stage of silica formation. The data clearly show the power of time-resolved fluorescence anisotropy decay measurements for probing the growth of silica colloids and show that this method is useful for elucidating the mechanism of particle formation and growth in situ.
Introduction Most theoretical models for silica particle growth use tetraethyl orthosilicate (TEOS) as the precursor and propose the nucleation of primary silica particles (aggregates of silica cycles with a radius of 1-4.5 nm)1 and their fast evolution into secondary particles and higher order aggregates. However, the notion that the primary particles actually exist in solution as discrete, stable structures has been viewed with skepticism for many years, mainly because of the extremely fast kinetics of silica growth in aqueous solutions and the limitations in the methods used to follow in situ such fast processes occurring on the nanoscale.2 Primary silica particles in solution are initially assembled through relatively weak bonds, and their character should depend on the entrained solvent and reagents. Drying the sample, which is required for ultrastructural analysis, completely changes the constitution of the particles.3 In solution, temporal changes occur within the particles and, thus, the time resolution of a measurement method should be shorter than the time * To whom correspondence should be addressed. Phone: (905) 525-9140 (ext. 27033). Fax: (905) 527-9950. E-mail: brennanj@ mcmaster.ca. Internet: http://www.chemistry.mcmaster.ca/faculty/ brennan. † Department of Chemistry. ‡ Brockhouse Institute for Materials Research. (1) Geddes, C. D.; Birch, D. J. S. J. Non-Cryst. Solids 2000, 270, 191. (2) Boukari, H.; Lin, J. S.; Harris, M. T. J. Colloid Interface Sci. 1997, 194, 311. (3) Hiemenz, P. C.; Rajagopalan, R. Principles of Colloid and Surface Chemistry, 3rd ed.; Marcel Dekker: New York, 1997.
scale of the silica transformation under investigation. On the basis of these requirements, most methods for probing primary particle formation have been based on smallangle X-ray scattering (SAXS) techniques, which show the appearance of polymeric particles with a radius of ∼10 nm even at early times during the Sto¨ber silica synthesis.2 More recent work using synchrotron radiation sources in combination with stopped-flow analysis allowed the detection of structures that were ∼3 nm in radius within 65 s of the initiation of the sol-gel reaction.4 These particles rapidly coalesced to form larger particles. The use of SAXS techniques also allowed the simultaneous observation of distinct populations of subcolloidal and colloidal particles throughout zeolite syntheses, where the speciation of silica precursors and the growth of nonspherical “nanoslabs” are controlled by organic cations, solvents, and high temperatures.5-7 Geddes and Birch8-11 introduced a new approach for following particle evolution that exploits a relationship (4) Pontoni, D.; Narayanan, T.; Rennie, A. R. Langmuir 2002, 18, 56. (5) Watson, J. N.; Iton, L. E.; Keir, R. I.; Thomas, J. C.; Dowling, T. L.; White, J. W. J. Phys. Chem. B 1997, 101, 10094. (6) Kirschhock, C. E.; Ravishankar, R.; Verspeurt, F.; Grobet, P. J.; Jacobs, P. A.; Martens, J. A. J. Phys. Chem. B 1999, 103, 4965. (7) de Moor, P. E.; Beelen, T. P.; van Santen, R. A. J. Phys. Chem. B 1999, 103, 1639. (8) Birch, D. J. S.; Geddes, C. D. Phys. Rev. E 2000, 62, 2977. (9) Geddes, C. D.; Karolin, J.; Birch, D. J. S. J. Fluoresc. 2002, 12, 113. (10) Geddes, C. D.; Karolin, J.; Birch, D. J. S. J. Fluoresc. 2002, 12, 135. (11) Geddes, C. D. J. Fluoresc. 2002, 12, 343.
10.1021/la0495478 CCC: $27.50 © 2004 American Chemical Society Published on Web 06/12/2004
Formation and Growth Mechanisms of Silica Particles
between the rotational characteristics of entrapped fluorescent probes and the evolution of particle growth in silica-based sols. According to this theory, rotational correlation times can be subdivided into three classes: (i) picosecond time scale rotational correlation times that correspond to a probe that is free in solution, (ii) nanosecond scale rotational correlation times, which are often observed in the time-resolved fluorescence anisotropy (TRFA) decays of cationic probes such as rhodamine 6G (R6G) in silica sols, that correspond to dye molecules that are electrostatically bound to primary silica nanoparticles,11 and (iii) residual anisotropy (r∞) values that correspond to dye molecules that are rigidly bound to either secondary (radius, 4.5 nm) or higher order particles such as silica aggregates and thus rotate too slowly to cause fluorescence depolarization during the 1-10 ns emission lifetime of a typical fluorescent probe.11 Since the nanosecond rotational time is ascribed to the rotation of primary silica nanoparticles, this theory allows for the calculation of the real-time particle size (assuming a spherical particle shape) from the Stokes-Einstein-Debye (SED) equation12
φ ) ηV/kT
(1)
where φ is the rotational correlation time of the particlebound probe, η is the microviscosity, V is the particle volume, T is the temperature, and k is the Boltzmann constant. The fractional fluorescence of each of the three decay components also provides useful information on the proportion of large and small silica structures and on the amount of free probe in the system. In cases where the photophysical properties of the probe (absorptivity coefficient, emission wavelength, and quantum yield) do not change as a result of adsorption (as is the case for the R6G-silica system),11 the fractional fluorescence of each decay component corresponds to the relative concentration of the probe in each environment. As shown herein, this information can be used to follow the time-dependent evolution of particle formation from nucleation to primary particle growth to aggregation. Prior work by Geddes and Birch demonstrated that the real-time growth of silica nanoparticles in aqueous sodium silicate is too fast to follow the initial stages of nucleation and silica particle growth: in their system, fully formed primary silica particles were observed using a gated measurement method in as little as 7 s after the formation of the sol solution.13 The formation of primary particles could be detected in tetramethoxysilane (TMOS) sols. However, their growth occurred over a very limited range, from ∼0.8 to 1.1 nm, and thus, later stages in silica particle formation could not be directly observed by the nanoparticle metrology approach. While TRFA measurements suggest the existence of primary particles, the only direct evidence for the presence of primary silica particles (radius, ∼2 nm) was provided by cryogenic transmission electron microscopy (TEM) of a base-catalyzed TEOS gel.14 However, in a later report,15 it was recognized that the 2 nm radius particles seen upon drying were artifacts of the procedure used to dry the specimen and were not present in the liquid sol. Hence, even though the formation of primary silica particles was (12) Debye, P. Polar Molecule; Chemical Catalog Co.: New York, 1929. (13) Geddes, C. D.; Karolin, J.; Birch, D. J. S. J. Phys. Chem. B 2002, 106, 3835. (14) Bailey, J. K.; Nagase, T.; Broberg, S. M.; Mecartney, M. L. J. Non-Cryst. Solids 1989, 109, 198. (15) Bailey, J. K.; Mecartney, M. L. Colloids Surf. 1992, 63, 151.
Langmuir, Vol. 20, No. 14, 2004 5925
postulated many years ago16,17 and is currently considered in all the models describing silica particle growth from monomeric species, their presence in silica sols has never been directly visualized. Our group recently developed a diglycerylsilane (DGS) precursor that evolves glycerol as the byproduct of the initial hydrolysis reaction.18 This precursor shows rapid hydrolysis over a broad pH range (pH 5-11),19 removing the need to use multiple pH regimes to initiate hydrolysis and condensation. More importantly, the presence of glycerol acts to suppress the rate of silica condensation reactions, which are normally very fast in the aqueous environment. As a result, the colloidal particle growth rate is retarded.13,20 As shown herein, fluorescence anisotropy can be used to follow, using a nanoparticle metrology approach, the formation and growth of stable primary particles in sols derived from DGS. More importantly, the existence of such particles bearing stabilizing layers is directly visualized for the first time by tapping mode and phase contrast atomic force microscopy (AFM). We also show that the slow particle growth rate in DGS sols makes this system suitable for examining, by anisotropy decay measurements, the mechanisms of particle formation, growth, and aggregation from the earliest stages of silica oligomer formation. Experimental Section Chemicals. Rhodamine 6G (R6G) was obtained from Sigma (St. Louis, MO). Ludox AM-30, 30 wt % SiO2 (average particle radius, 6 nm; specific surface area, 220 m2‚g-1), was obtained from DuPont. Sodium silicate solution (SiO2, 27 wt %; NaOH, 14 wt %), tetramethyl orthosilicate, and Dowex 50WX8-100 ionexchange resin of analytical grade were purchased from Aldrich (Milwaukee, WI). Diglycerylsilane (DGS, Si(Gly)2; MW, 208.3) was synthesized as described elsewhere.19 All water was distilled and deionized using a Milli-Q Synthesis A10 water purification system. All other reagents were used without further purification. Procedures. Solutions calculated to produce ∼3 wt % colloidal silica suspensions were prepared from Ludox (AM-30; nominal particle radius, g6 nm;21 all samples were prepared and then filtered through a 0.45 µm Acrodisc filter); acidification of sodium silicate via reaction with Dowex resin;20 or acid-catalyzed hydrolysis of tetramethoxysilane (TMOS; prepared following the method of Narang22) or DGS, respectively, and were examined using steady-state and time-resolved anisotropy methods. DGS sols containing 0.2-3 wt % SiO2 were prepared by dissolving different amounts (0.3-2.4 g) of finely ground DGS powder in 20 mL of borate buffer (20 mM, pH 8.2 or 9.2) that was previously cooled to 4 °C. The borate buffer contained R6G at a concentration of 1 µM. After ∼1 min of strong vortexing or sonication, the DGS suspension was filtered through a 0.45 µm membrane filter immediately before the steady-state or time-resolved anisotropy measurements were done. The steady-state fluorescence anisotropy measurements were performed using a SLM 8100 spectrofluorimeter (Spectronic Instruments, Rochester, NY), as described elsewhere,20,23 using λex ) 527 nm and λem ) 551 nm. The time-resolved fluorescence (16) Iler, R. K. The Chemistry of Silica: Solubility, Polymerization, Colloid and Surface Properties, and Biochemistry; Wiley-Interscience Publication: New York, 1979. (17) Brinker, C. J.; Scherer, G. W. Sol-Gel Science: The Physics and Chemistry of Sol-Gel Processing; Academic Press: San Diego, CA, 1990. (18) Brook, M. A.; Chen, Y.; Guo, K.; Zhang, Z.; Jin, W.; Deisingh, A.; Brennan, J. D. J. Sol.-Gel Sci. Technol., in press. (19) Chen, Y.; Brennan, J. D.; Brook, M. A. U.S. Provisional Patent 60/384,084, Aug 23, 2002. (20) Tleugabulova, D.; Zhang, Z.; Brennan, J. D. J. Phys. Chem. B 2003, 107, 10127. (21) Ludox Colloidal Silica, Properties, Uses, Storage and Handling; data sheet; Du Pont: 1987. (22) Narang, U.; Wang, R.; Prasad, P. N.; Bright, F. V. J. Phys. Chem. 1994, 98, 17. (23) Zheng, L.; Reid, W. R.; Brennan, J. D. Anal. Chem. 1997, 69, 3940.
5926
Langmuir, Vol. 20, No. 14, 2004
Tleugabulova et al.
intensity and anisotropy decay data of R6G were acquired in the time domain using an IBH 5000U time-correlated single photon counting fluorimeter, as described elsewhere.20 In all cases, the intensity decay data could be fit to a single decay time. The anisotropy decay was best fit to a two-component hindered rotor model according to the following equation:11
r(t) ) fr0 exp(-t/φ1) + (1 - f - g)r0 exp(-t/φ2) + gr0 (2) where φ1 reflects rapid rotational motion associated with rotation of the free probe in solution, φ2 reflects slow rotational reorientation of the probe bound to silica nanoparticles, f is the fraction of free probe in solution, (1 - f - g) is the fraction of probe bound to silica, g is the fraction of probe that is rigidly bound to larger particles that rotate more slowly than can be measured with the R6G probe (φ > 60 ns), and r0 is the limiting anisotropy. In some cases, the value of gr0 is denoted as r∞, the residual anisotropy. Fits were considered acceptable if the reduced chi-squared value (χ2R) was close to 1.0 and the residuals showed a random pattern. Failure to include a residual anisotropy term generally led to inferior fits (χ2R > 1.2). Atomic Force Microscopy. AFM imaging was performed using a Digital Instruments NanoScope IIIa Multimode instrument. One drop of a 1/200 diluted silica sol containing 0.015 wt % SiO2 was placed onto an electronics grade silicon substrate (mean roughness,
