Article pubs.acs.org/cm
Preparation of Size-Controlled Monodisperse Colloidal Mesoporous Silica Nanoparticles and Fabrication of Colloidal Crystals Eisuke Yamamoto,† Masaki Kitahara,† Takuya Tsumura,† and Kazuyuki Kuroda*,†,‡ †
Department of Applied Chemistry, Faculty of Science, Engineering, Waseda University, Ohkubo 3-4-1, Shinjuku-ku, Tokyo 169-8555, Japan ‡ Kagami Memorial Research Institute for Materials Science, Technology, Waseda University, Nishiwaseda 2-8-26, Shinjuku-ku, Tokyo 169-0051, Japan S Supporting Information *
ABSTRACT: Although mesoporous silica particles are useful building blocks for colloidal crystals, mesoporous silica nanoparticles smaller than 100 nm with sufficient monodispersity and colloidal stability to enable thermodynamic assemblies have not been reported. Here, we report that highly monodisperse colloidal mesoporous silica nanoparticles (CMS) can be prepared by combining the preparation of colloidal mesoporous silica nanoparticles with a shortened nucleation time. The nanoparticles exhibited a uniform shape and relatively smooth surface because an undesirable aggregate dispersion process was avoided. In addition, the diameter of the nanoparticles was controlled by seed-growth without spontaneous nucleation, which enabled the investigation of fundamental CMS properties. Using monodisperse CMS, the dependence of the ζ-potential of CMS on the diameter was revealed. Colloidal crystals composed of mesoporous silica nanoparticles were fabricated by drying the colloidal solution. This is the first report regarding the fabrication of colloidal crystals composed of mesoporous silica nanoparticles with a small particle size.
■
INTRODUCTION Colloidal crystals prepared by the assembly of colloidal particles have been studied for various applications such as photonic crystals, sensors, templates, and catalysts because of their threedimensionally (3D) periodic refractive indices and interstitial 3D ordered pores.1 Controlling the particle size of colloidal particles is essential for various applications because the properties of colloidal crystals strongly depend on particle size. In particular, colloidal crystals composed of nanoparticles (defined here as particles smaller than 100 nm) have attracted much attention as masks for colloidal lithography,2 templates for the preparation of meso(macro)porous materials3 that are useful for various applications,4 and photonic crystals with band gaps in the UV region.5 In addition, colloidal crystals composed of nanoparticles can be used as a template for the preparation of anisotropically grown nanostructured gold, which has not been achieved by conventional templating.6 To exploit the unique features of such colloidal crystals more effectively, it is important to control the particle size, composition, structure, and morphology of the nanoparticles. Particles used for the preparation of colloidal crystals are generally nonporous silica or polymers.1k On the other hand, mesoporous silica particles are attractive components of colloidal crystals because of their tuned functionalities, such as refractive indices, by using uniform mesopores, high surface areas, and high pore volumes.7 In fact, colloidal crystals composed of mesoporous silica particles form hierarchical © 2014 American Chemical Society
porous structures possessing an interstitial space between particles and mesopores within particles, which enables us to utilize the features of colloidal crystals more effectively.8 For example, the refractive index can be controlled in a wider range because large amounts of guest species can be introduced into the interstitial space and/or ordered mesopores.8h,i In addition, when mesoporous silica colloidal crystals are utilized as templates, hierarchical nanostructured materials reflecting both interstices and mesopores can be obtained. Colloidal crystals composed of mesoporous silica nanoparticles should have great potential for various applications. However, to the best of our knowledge, there are no papers on the formation of colloidal crystals composed of mesoporous silica nanoparticles. Three papers have reported the fabrication of 3D colloidal crystals composed of mesoporous silica particles, but the reported particle sizes were 550,7e 150,7c and 250−1500 nm.7d The smallest size is still larger than 100 nm and the surfaces of 150 nm particles are rough and somewhat uncontrollable.7c Therefore, it is essential to develop a novel method to fabricate colloidal crystals composed of mesoporous silica nanoparticles. To prepare colloidal crystals, particles must meet the following requirements: (1) monodispersity, defined here as Received: February 20, 2014 Revised: April 4, 2014 Published: April 11, 2014 2927
dx.doi.org/10.1021/cm500619p | Chem. Mater. 2014, 26, 2927−2933
Chemistry of Materials
Article
Scheme 1. Schematic for the Preparation of Monodisperse Colloidal Mesoporous Silica Nanoparticles and Fabrication of Colloidal Crystals
However, the particle size prepared by this method is not uniform.10i Therefore, in order to overcome the issue, we decreased the number of formed particles by adding a small amount of silica, thus shortening the nucleation time which is strongly related to particle size. Thus, particles with a uniform size were prepared. In addition, we demonstrate the sizecontrol of monodisperse mesoporous silica nanoparticles via a seed-growth method.14 Control of the particle size is essential because the size greatly affects the features of colloidal crystals. Furthermore, an amphoteric ion-exchange resin was immersed in the colloidal solution of the nanoparticles to increase the colloidal stability,15 because the colloidal stability of nanoparticles obtained using dialysis was insufficient to entirely form colloidal crystals. Colloidal crystals and composed of mesoporous silica nanoparticles smaller than 100 nm were successfully formed for the first time by drying colloidal solutions after immersion in an ion-exchange resin.
uniform size and shape of the particles and (2) colloidal stability to enable thermodynamically favorable assemblies.1k,9 Although there have been many reports on the preparation of uniform-sized mesoporous silica nanoparticles,10 the shapes are not uniform and the monodispersity of the particles is relatively low in almost all reports, with the exception of the following three papers: Huo et al. reported the preparation of monodisperse colloidal mesoporous silica nanoparticles (CMS) by controlling the pH of the solution.10b Yu et al. reported the preparation of CMS by retaining ethoxy groups on the surface of particles.10a Chen et al. also succeeded in preparing monodisperse mesoporous silica nanoparticles with a standard deviation of the diameter under 10%.10c However, to date, the preparation of colloidal crystals with ordered arrays using nanoparticles has not yet been achieved probably because the colloidal stability of the nanoparticles is insufficient to form colloidal crystals. In general, the colloidal stability of particles decreases as the size decreases.11 Therefore, the fabrication of colloidal crystals composed of nanoparticles is more challenging than the fabrication of colloidal crystals composed of macrosized particles. Accordingly, fabrication of colloidal crystals requires the preparation of mesoporous silica nanoparticles with sufficient monodispersity and colloidal stability to enable thermodynamically favorable assemblies. Here, we report the preparation of monodisperse mesoporous silica nanoparticles with high colloidal stability and the subsequent fabrication of colloidal crystals composed of nanoparticles. The overall scheme of the preparation is shown in Scheme 1. We have combined a preparative method for colloidal mesoporous silica nanoparticles reported previously10i and a shortened nucleation time to yield a narrow particle-size distribution.12 In the preparative method, because of the presence of a large amount of C16TMABr, highly dispersed and uniformly shaped colloidal mesostructured silica nanoparticles can be obtained.13 In addition, the surfactants of colloidal mesostructured silica nanoparticles can be removed by dialysis without the concomitant formation of aggregates. Accordingly, the surface of nanoparticles is relatively smooth, which is suitable for the fabrication of colloidal crystals.
■
EXPERIMENTAL SECTION
Materials. All materials were used without further purification. Hexadecyltrimethylammonium bromide (C16TMABr), triethanolamine (TEA) as a base catalyst, and acetic acid were purchased from Wako Pure Chem. Ind., Ltd. Tetraethoxysilane (TEOS: Si(OC2H5)4) and tetrapropoxysilane (TPOS: Si(OC3H7)4) were purchased from Tokyo Kasei Co., Ltd. Amberlite MB-1 (Dow Chemical Co.) was used as an amphoteric ion-exchange resin. Characterization. Dynamic light scattering (DLS) measurements were conducted on a HORIBA Nano Partica SZ-100-S at 25 °C. Transmission electron microscope (TEM) images were obtained on a JEOL JEM-2010 microscope operating at 200 kV. Samples for TEM measurements were dropped and dried on a carbon-coated microgrid (Okenshoji Co.). Mean particle sizes and standard deviations were obtained by measuring the size of 150 nanoparticles in the TEM images. X-ray diffraction (XRD) patterns of dried powder samples were obtained on Rigaku Ultima IV with Fe Kα radiation (40 kV, 30 mA). Nitrogen gas adsorption−desorption measurements were performed with an Autosorb-1 instrument (Quantachrome Instruments) at −196 °C. Samples were preheated at 120 °C for 24 h under vacuum. The pore-size distributions were roughly evaluated using the BJH (Barrett−Joyner−Halenda) method. CHN analysis data were obtained with PerkinElmer 2400 Series II. Thermogravimetry (TG) 2928
dx.doi.org/10.1021/cm500619p | Chem. Mater. 2014, 26, 2927−2933
Chemistry of Materials
Article
curves were obtained on a RIGAKU Thermo Plus 2 instrument under a dry air flow at a heating rate of 10 °C min−1 up to 900 °C. ζPotential measurements were conducted with Otsuka Electronics ELSZ-1 at 20 °C using the Smoluchowski equation.16 This equation is normally applied to estimate the ζ-potential of relatively large colloidal dense particles for quantitative discussions. However, very small mesostructured nanoparticles are utilized in the present study, and there are no theoretical methods to calculate the ζ-potentials of “mesostructured” and “nano-sized” particles exactly. Therefore, we used the ζ-potential calculated by the Smoluchowski equation for qualitative discussions. ζ-Potential values may be estimated as slightly larger than the true values with the decrease in the diameter of the nanoparticles, because the strain of the electrical field around the surfaces of nanoparticles in electrophoresis was ignored.16 Preparation of Monodisperse Colloidal Mesostructured Silica Nanoparticles As Seed Particles. A precursor solution was prepared according to literature.10i C16TMABr (0.66 g) and TEA (0.14 g) were dissolved in 80 mL of water and the solution was stirred at exactly 80 °C for 1 h. Notably, the precise control of the temperature was quite important in controlling the number of nuclei. Next, TEOS (0.156 mL) was added to the solution and the mixture was stirred vigorously at 80 °C for 1 h. After stirring, the mixture was cooled to 30 °C slowly. The temperature of the solution is important in the preparation of monodisperse particles. The colloidal solution is denoted as CMS-0-as, where “as” means “as-synthesized” and “0” means that another silica source was not added for the growth of the nuclei. Please note that CMS-0-as does not represent colloidal particles but represents the solution containing colloidal particles. Preparation of Monodisperse Colloidal Mesostructured Silica Nanoparticles by Seed-Growth Method. TPOS (0.464 mL) was added to CMS-0-as and the colloidal solution was stirred at 30 °C for 1 d. This procedure was repeated 1 to 4 times. The samples are denoted as CMS-x-as, where “x” indicates the cycle number of the procedure. All colloidal solutions were filtered by filter paper (No. 5c) to remove a very small amount of solid adhered to the surface of the containers and impurities such as dust. Therefore, almost all silica species were recovered as colloids. Extraction of Surfactants from Mesostructured Nanoparticles. A sample of CMS-x-as (50 mL) was transferred into a dialysis membrane tube composed of cellulose (molecular weight cutoff 12000−14000) and was dialyzed for 12 h against a mixture (250 mL) of 2 M acetic acid and ethanol (1:1, v/v) to remove C16TMABr. This process was repeated five times. Next, the tube that contained CMS was immersed in deionized water to remove acetic acid and ethanol and was repeated four times. These samples are denoted as CMS-x-dia, where “dia” indicates “after dialysis”. Fabrication of Colloidal Crystals Composed of Mesoporous Silica Nanoparticles. To increase the colloidal stability of CMS-xdia, Amberlite was immersed into CMS-x-dia and the colloidal solution was left for 2 d at room temperature. These samples are denoted as CMS-x-DI, where “DI” means “after deionized”. Finally, colloidal crystals were obtained by drying CMS-x-DI slowly at room temperature on a plastic boat. These samples are denoted as Crystal-x-DI. The samples were easily removed from the boat for characterization. To estimate the effect of the immersion of Amberlite, colloidal crystals were obtained by drying CMS-x-dia slowly at room temperature on a plastic boat without the addition of Amberlite. These samples are denoted as Crystal-x-dia.
CMS-0-as retained the dispersion of primary nanoparticles. From the TEM image (Figure 1c), the mean diameter of the particles was estimated to be ca. 40 nm. The hydrodynamic diameter was larger than the particle diameter estimated by TEM because of the presence of a hydration layer and diffuse electrical double layer. The standard deviation of the diameter was calculated to be 8.9% from TEM, which indicated that highly monodisperse nanoparticles were successfully obtained. While the standard deviation of the nanoparticles prepared according to a literature17 was 35%, that of the nanoparticles in CMS-0-as was 8.9%. (Figure S1, Supporting Information (SI)) The higher monodispersity of the nanoparticle in CMS-0-as was attributed to the reduction of the particle number. In general, the broadness of the particle diameter distribution depends on the nucleation time. A longer nucleation time promotes a different growth time of each particle, resulting in the formation of particles with different sizes,14a,b as shown in Figure 2. In the case of nanoparticles, the preparation of monodisperse particles is challenging because their particle number is typically large, which leads to a longer nucleation time. Therefore, a significant reduction in the particle number is essential for the preparation of monodisperse nanoparticles. In this report, we added a smaller amount of silica to reduce the particle number than those reported previously. In fact, even though the amount of silica was extremely low, CMS-0-as contained larger particles than nanoparticles prepared using larger amounts of silica,17 which demonstrated the effect of the reduced particle number. Therefore, the standard deviation of the diameter of nanoparticle in CMS-0-as was lower than that of nanoparticles reported previously. This result suggests that our preparative method is quite effective for the preparation of monodisperse colloidal mesostructured nanoparticles. The mesostructured silica nanoparticles obtained before dialysis were therefore used as the seeds, because the nanoparticles were dissolved when dialysis was carried out, owing to the very small amount of silica. Preparation of Monodisperse Mesoporous Silica Nanoparticles by Seed-Growth Method for Fabrication
RESULTS AND DISCUSSION Preparation of Monodisperse Colloidal Mesostructured Silica Nanoparticles As Seed Particles. Figure 1a shows the appearance of the CMS-0-as. CMS-0-as was almost transparent in the visible light region and no precipitation was observed. To measure the DLS of CMS-0-as, CMS-0-as was concentrated by centrifugation using a centrifugal filter, because CMS-0-as was too dilute to be measured. The hydrodynamic diameter distribution calculated from DLS data showed only one peak at ca. 66 nm (Figure 1b). These results indicated that
Figure 2. Effect of long nucleation time. Longer nucleation time leads to broader particle diameter distribution.
Figure 1. (a) Appearance of CMS-0-as, (b) hydrodynamic diameter distribution of CMS-0-as measured by dynamic light scattering, and (c) TEM image of CMS-0-as.
■
2929
dx.doi.org/10.1021/cm500619p | Chem. Mater. 2014, 26, 2927−2933
Chemistry of Materials
Article
of Colloidal Crystals. Monodisperse colloidal mesostructured silica nanoparticles with different sizes were prepared by the seed-growth method using the nanoparticles in CMS-0-as as seeds. All CMS-x-as (x = 1−4) were nearly transparent in the visible light region and no precipitates were observed (Figure S2, SI). The hydrodynamic diameter distribution calculated from DLS showed only one peak at ca. 86, 106, 122, and 143 nm, respectively (Figure S3, SI). The TEM images of CMS-x-as (x = 1, 2, 3, and 4) showed that the nanoparticles had uniform spherical shapes and relatively smooth surfaces. Moreover, the mean diameter of the nanoparticles was 73, 82, 92, and 108 nm, respectively (Figure S4, SI). As described above, it is reasonable that the hydrodynamic diameters of the particles were larger than those estimated by TEM. The standard deviations of the diameters were calculated to be 8.0%, 6.9%, 7.0%, and 5.6% for CMS-x-as (x = 1, 2, 3, and 4), respectively, by TEM. In addition, TEM images showed that CMS-x-as had a mesoscale periodicity of ca. 4 nm. These results indicated that all samples retained the dispersion of the primary nanoparticles during the seed-growth process. Monodisperse colloidal mesoporous silica nanoparticles were obtained using dialysis. All CMS-x-dia (x = 1−4) were nearly transparent in the visible light region and no precipitates were observed (Figure S5, SI). The hydrodynamic diameters of the nanoparticles in CMS-x-dia (x = 1, 2, 3, and 4) (Figure S6, SI) were almost the same as those of the nanoparticles in CMS-x-as (x = 1, 2, 3, and 4). In addition, TEM images of CMS-x-dia (x = 1, 2, 3, and 4) (Figure S7, SI) showed that the mean diameter, standard deviation of the particle diameter distribution, shape, surface structure, and mesostructure of nanoparticle in CMS-x-dia were almost the same as those of nanoparticle in CMS-x-as (x = 1, 2, 3, and 4). These results indicated that all samples retained the primary nanoparticle dispersion during dialysis and that the nanoparticles were not dissolved during dialysis because the amount of silica in CMS-x-as was higher than that of CMS-0-as. CMS-x-DI were prepared by immersing the ion-exchange resin into the CMS-x-dia. The removal of surfactants was confirmed by CHN and TG analyses (Figure S8, SI). The nitrogen content of CMS-4-DI was under the detection limit of CHN, and the carbon content was 0.3 mass%. The TG curve of CMS-4-DI displayed two weight losses at 25−150 °C and higher than 150 °C, which corresponded to the desorption of water and gradual dehydration of the silanol groups, respectively. These results suggested that almost all surfactants were removed during dialysis and immersion. All CMS-x-DI (x = 1−4) were nearly transparent in the visible light region and no precipitates were observed (Figure 3). The hydrodynamic diameters of CMS-x-DI (x = 1, 2, 3, and 4) were almost the same as those of the nanoparticles in CMS-x-as (x = 1, 2, 3, and 4), respectively (Figure 4). In addition, TEM images of CMS-xDI (x = 1, 2, 3, and 4) showed that the mean diameter, standard deviation of the particle diameter distribution, shape, surface structure, and mesostructure of nanoparticle in CMS-xDI were almost the same as those of nanoparticle in CMS-x-as (x = 1, 2, 3, and 4) (Figure 5). These results indicated that the immersion of CMS-x-dia in the ion-exchange resin did not affect the structure and size of the nanoparticles. As shown above, the size of nanoparticles in CMS-x-as was successfully increased without any concomitant formation of spontaneously nucleated particles by the seed-growth method. The precise control of the size could be attributed to the small
Figure 3. Appearances of (a) CMS-1-DI, (b) CMS-2-DI, (c) CMS-3DI, and (d) CMS-4-DI.
Figure 4. Hydrodynamic diameter distributions of (a) CMS-1-DI, (b) CMS-2-DI, (c) CMS-3-DI, and (d) CMS-4-DI, measured by dynamic light scattering.
Figure 5. TEM images of (a) CMS-1-DI, (b) CMS-2-DI, (c) CMS-3DI, and (d) CMS-4-DI.
number of particles in the colloidal solution and the low hydrolysis rate of the silica sources, as explained below. 2930
dx.doi.org/10.1021/cm500619p | Chem. Mater. 2014, 26, 2927−2933
Chemistry of Materials
Article
In general, to grow particles without spontaneous nucleation, the concentration of the solute should be kept below the critical nucleation concentration. A highly concentrated C16TMABr solution, which is necessary for preparation of uniformly shaped colloidal particles,13 should promote the nucleation of the particles by decreasing the critical nucleation concentration. In our preliminary study on the growth of nanoparticles by the seed-growth method in a highly concentrated C16TMABr solution, a large number of spontaneous nuclei were observed during the seed-growth process. On the other hand, the precise control of the size of the nanoparticles studied here was achieved without spontaneous nucleation. The suppression of spontaneous nuclei should be due to the decreased supply rate of silicate, which is caused by the following three factors: (1) The temperature of the colloidal solutions was lowered when the silica was added; (2) the hydrolysis rate of TPOS is relatively slow;17 (3) the amount of added TPOS was small. Nanoparticles with precisely controlled sizes are useful not only for the fabrication of colloidal crystals but also in various applications and in investigations of the fundamental properties of CMS. Fabrication of Colloidal Crystals Composed of Mesoporous Silica Nanoparticles. Colloidal crystals composed of mesoporous silica nanoparticles (Crystal-x-DI) were prepared by drying CMS-x-DI. Ordered arrays of nanoparticles were mainly observed in the SEM of Crystal-1DI and Crystal-2-DI (Figure 6a and b), though disordered colloidal crystals were partially obtained. The ordered arrays of nanoparticles were entirely observed in the SEM images of Crystal-3-DI and Crystal-4-DI (Figure 6c and d and Figure S9, SI). Fourier transformed images derived from the corresponding SEM images showed spots that could be assigned to an ordered hexagonal array (Figure S10, SI).These results indicated the formation of colloidal crystals. However, the mesostructure of the nanoparticles was not clearly observed because nanoparticles were heavily charged owing to their low electric conductivity. Therefore, the mesostructure of the nanoparticles was confirmed by XRD and N2 adsorption−desorption measurements, as shown below. The XRD patterns of Crystal-X-DI (Figure S11, SI) indicated the formation of a mesostructure derived from the micelle of C16TMABr. The d values were 4.9, 5.2, 5.8, and 6.0 nm for
Crystal-x-DI (x = 1, 2, 3, and 4), respectively. This result suggested that the mesostructure of the nanoparticles was retained after increasing the size of the particles. In the N2 adsorption−desorption isotherms, two steep increases were observed at P/P0 = ca. 0.4 and 0.9, respectively (Figure 7). The pore-size distributions of Crystal-x-DI calculated from the desorption isotherms by the BJH method showed two distinct peaks at ca. 4 and 20 nm, respectively (Figure 8). The smaller pore size was consistent with that estimated from the TEM observation of CMS-x-DI (Figure 4). These results indicated that the nanoparticles themselves had mesopores. The larger pore size should be ascribed to interparticle mesopores and the peaks around 20 nm were steeper than those observed for mesoporous silica nanoparticles with lower colloidal stabilities and monodispersities.10 In addition, the size of the interparticle mesopores decreased slightly as the particle size decreased. These results strongly suggested that the interparticle mesopores were uniform, and colloidal crystals were entirely ordered in all Crystal-x-DI (x = 1, 2, 3, and 4). This is the first report of the preparation of colloidal crystals composed of mesoporous silica nanoparticles smaller than 100 nm. Effect of Ion-Exchange Resin Immersion on the Fabrication of Colloidal Crystals. CMS-x-dia were used for the fabrication of colloidal crystals (Crystal-x-dia). The regularities of the arrays were obviously lower than those of Crystal-x-DI. Although ordered colloidal crystals were observed in the SEM of Crystal-x-dia, disordered arrays of Crystal-x-dia were also observed (Figure S12, SI). In addition, the packing quality of Crystal-1-dia and Crystal-2-dia seemed to be lower than that of Crystal-1-DI and Crystal-2-DI. The difference in the regularity of the arrays was also suggested from the N2 adsorption−desorption measurements (Figures S13 and S14, SI). The peaks assigned to the interparticle pores in the poresize distribution of Crystal-x-dia were slightly broader than those of Crystal-x-DI. In addition, the pore size of Crystal-1-dia and Crystal-2-dia were larger than Crystal-1-DI and Crystal-2DI. These results also suggested that the regularities of the arrays of Crystal-x-dia were lower. Because the particle size of Crystal-x-dia was almost the same as that of Crystal-x-DI, the difference in the regularity of the arrays should be caused by the
Figure 6. SEM images of (a) Crystal-1-DI, (b) Crystal-2-DI, (c) Crystal-3-DI, and (d) Crystal-4-DI.
Figure 7. N2 adsorption−desorption isotherms (a) Crystal-1-DI, (b) Crystal-2-DI, (c) Crystal-3-DI, and (d) Crystal-4-DI. 2931
dx.doi.org/10.1021/cm500619p | Chem. Mater. 2014, 26, 2927−2933
Chemistry of Materials
Article
Figure 9. Appearance of CMS-4-DI when the ion-exchange resin was immersed.
ionic concentration of the colloidal solutions should be nearly equal because the lengths and numbers of dialysis and deionization processes were identical. Consequently, the tendency indicates that the surface electrical potential decreased as the diameter of the nanoparticles decreased. Such a relationship would be explained by the presence of the silanolate groups inside the nanoparticles. While the silanolate groups of nonporous silica particles are located on the surface of particles, mesoporous silica nanoparticles have silanolate not only on their surfaces but also inside of them. Therefore, silanolates located inside nanoparticles would cause the variation in the surface potential depending on the diameter. In general, the colloidal stability of particles decreases as the particle diameter decreases when the surface electrical potential is constant because the decrease in the electrical repulsion forces exceeds the decrease in van der Waals attraction as particles size decreases.18 In addition, the colloidal stability also decreases as the surface potential decreases even though the particle size is constant. Therefore, the decrease of the colloidal stability of the mesoporous silica nanoparticles can be attributed to both of these factors. This finding was obtained using our method to prepare monodisperse CMS and by controlling the diameter without spontaneous nucleation. It reveals for the first time that the ζpotential of mesoporous silica nanoparticles depends on the diameter. This dependency is important for further investigations of interparticle forces of porous nanoparticles, such as van der Waals interactions and hydration forces.18 Information on the interparticle forces of mesoporous silica nanoparticles is important for various applications using colloidal dispersions, such as drug delivery systems and preparation of membranes composed of nanoparticles.
Figure 8. Pore diameter distributions of (a) Crystal-1-DI, (b) Crystal2-DI, (c) Crystal-3-DI, and (d) Crystal-4-DI.
difference in the colloidal stabilities of the nanoparticles. Therefore, we measured the ζ-potential of CMS-x-dia and CMS-x-DI to evaluate their colloidal stabilities. The measured ζ-potential values are given in Table 1. The ζ-potentials of CMS-x-DI were obviously increased compared to those of CMS-x-dia From these results, it is clear that the immersing procedure increased the ζ-potential. In colloidal solutions, ionic species such as acetic acid and/or C16TMA cations may shield the electrical charge of the silanolate groups on particles, which decreases the ζ-potential.15 Therefore, a decreased amount of ionic species should result in an increased ζ-potential. In general, an ion-exchange resin can remove such ions in solution even though the amount of ions is quite small. Accordingly, an increased ζ-potential by immersing Amberlite into the colloidal solution can be attributed to the decreased amount of ionic species that were not removed by dialysis. In addition, iridescence was observed after the immersion of the ion-exchange resin (Scheme 1 and enlarged photograph in Figure 9), which strongly supported the notion that the increased ζ-potential was due to the decrease in ionic species. Effect of the Size of Mesoporous Silica Nanoparticles on the Fabrication of Ordered Colloidal Crystals. Although CMS-1-dia and CMS-2-dia had nanoparticles with sufficient monodispersity to assemble into ordered colloidal crystals, disordered arrays were observed for Crystal-1-dia and Crystal-2-dia. Similarly, disordered arrays were partially observed for Crystal-1-DI and Crystal-2-DI, though CMS-1DI and CMS-2-DI had nanoparticles with sufficient monodispersity. These results indicated that the colloidal stability of the nanoparticles decreased as the diameter of the CMS decreased. Because the relationship between the colloidal stability of mesoporous silica nanoparticles and their diameter had not been investigated thus far, we compared the ζpotentials of larger particles with those of smaller particles. Interestingly, the ζ-potentials of CMS-x-dia and CMS-x-DI decreased as the diameter of the nanoparticles decreased. The
■
CONCLUSION Monodisperse colloidal mesoporous silica nanoparticles were prepared. The monodispersity of the nanoparticles was attributed to the shortened nucleation time. The nanoparticles had a uniform shape and relatively smooth surface because dispersion was avoided. The diameter of the nanoparticles was controlled using a seed-growth method without any spontaneous nucleation, and the ζ-potential of the nanoparticles depended on their diameter. Colloidal crystals composed of mesoporous silica nanoparticles smaller than 100 nm were prepared for the first time by drying the obtained colloidal solutions. This novel material will contribute to the control of the unique features of colloidal crystals composed of nanoparticles. The size-controlled monodisperse colloidal mesoporous silica nanoparticles reported here may be useful for the
Table 1. ζ-Potentials of CMS-x-dia and CMS-x-DI
CMS-x-dia CMS-x-DI
x=1
x=2
x=3
x=4
−8 mV −16 mV
−13 mV −20 mV
−17 mV −26 mV
−23 mV −30 mV 2932
dx.doi.org/10.1021/cm500619p | Chem. Mater. 2014, 26, 2927−2933
Chemistry of Materials
Article
A.; Yakovlev, S. A.; Kirilenko, D. A.; Kukushkina, Y. A.; Nashchekin, A. V.; Sitnikova, A. A.; Yagovkina, M. A.; Golubev, V. G. Nanotechnology 2013, 24 (15), 155601. (e) Yano, K.; Fukushima, Y. J. Mater. Chem. 2004, 14 (10), 1579. (8) (a) Murai, S.; Yao, S.; Nakamura, T.; Kawamoto, T.; Fujita, K.; Yano, K.; Tanaka, K. Appl. Phys. Lett. 2012, 101 (15), 151121. (b) Kelly, T. L.; Yamada, Y.; Schneider, C.; Yano, K.; Wolf, M. O. Adv. Funct. Mater. 2009, 19 (23), 3737−3745. (c) Nakamura, T.; Yamada, Y.; Yamada, H.; Yano, K. J. Mater. Chem. 2009, 19 (37), 6699. (d) Nakamura, T.; Yamada, Y.; Yano, K. J. Mater. Chem. 2007, 17 (35), 3726. (e) Nakamura, T.; Yamada, Y.; Yano, K. Microporous Mesoporous Mater. 2009, 117 (1−2), 478−485. (f) Wang, Y.; Caruso, F. Adv. Funct. Mater. 2004, 14 (10), 1012−1018. (g) Yamada, H.; Nakamura, T.; Yamada, Y.; Yano, K. Adv. Mater. 2009, 21 (41), 4134−4138. (h) Yamada, Y.; Nakamura, T.; Ishi, M.; Yano, K. Langmuir 2006, 22 (6), 2444−2446. (i) Yamada, Y.; Nakamura, T.; Yano, K. Langmuir 2008, 24 (6), 2779−2784. (9) von Freymann, G.; Kitaev, V.; Lotsch, B. V.; Ozin, G. A. Chem. Soc. Rev. 2013, 42 (7), 2528−2554. (10) (a) Yu, M.; Zhou, L.; Zhang, J.; Yuan, P.; Thorn, P.; Gu, W.; Yu, C. J. Colloid Interface Sci. 2012, 376 (1), 67−75. (b) Qiao, Z.-A.; Zhang, L.; Guo, M.; Liu, Y.; Huo, Q. Chem. Mater. 2009, 21 (16), 3823−3829. (c) Chen, C.-N.; Lin, H.-P.; Tsai, C.-P.; Tang, C.-Y. Chem. Lett. 2004, 33 (7), 838−839. (d) Chen, Q.; Han, L.; Gao, C.; Che, S. Microporous Mesoporous Mater. 2010, 128 (1−3), 203−212. (e) He, Q.; Cui, X.; Cui, F.; Guo, L.; Shi, J. Microporous Mesoporous Mater. 2009, 117 (3), 609−616. (f) Kim, T.-W.; Chung, P.-W.; Lin, V. S. Y. Chem. Mater. 2010, 22 (17), 5093−5104. (g) Lu, F.; Wu, S. H.; Hung, Y.; Mou, C. Y. Small 2009, 5 (12), 1408−1413. (h) Möller, K.; Kobler, J.; Bein, T. Adv. Funct. Mater. 2007, 17 (4), 605−612. (i) Urata, C.; Aoyama, Y.; Tonegawa, A.; Yamauchi, Y.; Kuroda, K. Chem. Commun. 2009, 34, 5094−5096. (j) Xu, H.; Li, Y.; Yang, Z.; Ding, Y.; Lu, C.; Li, D.; Xu, Z. Int. J. Appl. Ceram. Technol. 2012, 9 (6), 1112−1123. (k) Yokoi, T.; Karouji, T.; Ohta, S.; Kondo, J. N.; Tatsumi, T. Chem. Mater. 2010, 22 (13), 3900−3908. (11) Huang, Y.; Pemberton, J. E. Colloids surf. A 2011, 377 (1−3), 76−86. (12) Huang, Y.; Pemberton, J. E. Colloids surf. A 2010, 360 (1−3), 175−183. (13) Yamada, H.; Urata, C.; Higashitamori, S.; Aoyama, Y.; Yamauchi, Y.; Kuroda, K. ACS Appl. Mater. Interfaces 2014, 6, 3491−3500. (14) (a) Chang, S. M.; Lee, M.; Kim, W.-S. J. Colloid Interface Sci. 2005, 286 (2), 536−542. (b) Giesche, H. J. Eur. Ceram. Soc. 1994, 14 (3), 205−214. (c) Watanabe, R.; Yokoi, T.; Kobayashi, E.; Otsuka, Y.; Shimojima, A.; Okubo, T.; Tatsumi, T. J. Colloid Interface Sci. 2011, 360 (1), 1−7. (15) (a) Okubo, T. Polym. J. 2008, 40 (9), 882−890. (b) Okubo, T. Acc. Chem. Res. 1988, 21 (7), 281−286. (16) Smoluchowski, M. v. Z. Phys. Chem. 1917, 92, 129−168. (17) Yamada, H.; Urata, C.; Aoyama, Y.; Osada, S.; Yamauchi, Y.; Kuroda, K. Chem. Mater. 2012, 24 (8), 1462−1471. (18) (a) Kamiya, H.; Mitsui, M.; Takano, H.; Miyazawa, S. J. Am. Ceram. Soc. 2000, 83 (2), 287−293. (b) Bishop, K. J.; Wilmer, C. E.; Soh, S.; Grzybowski, B. A. Small 2009, 5 (14), 1600−30. (c) Verwey, G. W.; and Overbeek, J. T. G. The Theory of the Stability of Lyophobic Colloids. Elsevier: Amsterdam, The Netherlands, 1948.
fabrication of colloidal crystals, investigations of fundamental properties, and various applications.
■
ASSOCIATED CONTENT
S Supporting Information *
Additional figures (S1−S14) as described in the text. This material is available free of charge via the Internet at http:// pubs.acs.org.
■ ■
AUTHOR INFORMATION
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS The authors thank Prof. H. Kamiya (Tokyo University for Agriculture and Technology) for his kind advice on the analysis of ζ-potential. They also thank Mr. T. Matsuno (Waseda University) for his kind assistance with SEM measurements. This work was supported in part by Grant-in-Aid Exploratory Research.
■
REFERENCES
(1) (a) Aguirre, C. I.; Reguera, E.; Stein, A. Adv. Funct. Mater. 2010, 20 (16), 2565−2578. (b) Ge, J.; Yin, Y. Angew. Chem., Int. Ed. 2011, 50 (7), 1492−1522. (c) Göltner, C. G. Angew. Chem., Int. Ed. 1999, 38 (21), 3155−3156. (d) Lee, K.; Asher, S. A. J. Am. Chem. Soc. 2000, 122 (39), 9534−9537. (e) Marlow, F.; Muldarisnur; Sharifi, P.; Brinkmann, R.; Mendive, C. Angew. Chem., Int. Ed. 2009, 48 (34), 6212−6233. (f) Moon, J. H.; Yang, S. Chem. Rev. 2009, 110 (1), 547−574. (g) Stein, A. Microporous Mesoporous Mater. 2001, 44−45 (0), 227− 239. (h) Stein, A.; Schroden, R. C. Curr. Opin. Solid State Mater. Sci. 2001, 5 (6), 553−564. (i) Velev, O. D.; Lenhoff, A. M. curr. opin. colloid interface sci. 2000, 5 (1−2), 56−63. (j) Wang, J.; Zhang, Y.; Wang, S.; Song, Y.; Jiang, L. Acc. Chem. Res. 2011, 44 (6), 405−415. (k) Xia, Y.; Gates, B.; Yin, Y.; Lu, Y. Adv. Mater. 2000, 12 (10), 693− 713. (2) Yang, S.-M.; Jang, S. G.; Choi, D.-G.; Kim, S.; Yu, H. K. Small 2006, 2 (4), 458−475. (3) (a) Han, J.; Crespi, V. Phys. Rev. Lett. 2001, 86 (4), 696−699. (b) Egan, G. L.; Yu, J. S.; Kim, C. H.; Lee, S. J.; Schaak, R. E.; Mallouk, T. E. Adv. Mater. 2000, 12 (14), 1040−1042. (c) Gates, B.; Yin, Y.; Xia, Y. Chem. Mater. 1999, 11 (10), 2827−2836. (d) Johnson, S. A. Science 1999, 283 (5404), 963−965. (e) Kuroda, Y.; Yamauchi, Y.; Kuroda, K. Chem. Commun. 2010, 46 (11), 1827−1829. (f) Wang, Z.; Dornath, P.; Chang, C.-C.; Chen, H.; Fan, W. Microporous Mesoporous Mater. 2013, 181, 8−16. (4) (a) Chai, G. S.; Yoon, S. B.; Yu, J.-S.; Choi, J.-H.; Sung, Y.-E. J. Phys. Chem. B 2004, 108 (22), 7074−7079. (b) Kang, S.; Yu, J.-S.; Kruk, M.; Jaroniec, M. Chem. Commun. 2002, No. 16, 1670−1671. (c) Ma, G.; Yan, X.; Li, Y.; Xiao, L.; Huang, Z.; Lu, Y.; Fan, J. J. Am. Chem. Soc. 2010, 132 (28), 9596−9597. (d) Vu, A.; Li, X.; Phillips, J.; Han, A.; Smyrl, W. H.; Bühlmann, P.; Stein, A. Chem. Mater. 2013, 25 (21), 4137−4148. (e) Ye, L.; Wang, Y.; Chen, X.; Yue, B.; Chi Tsang, S.; He, H. Chem. Commun. 2011, 47 (26), 7389−7391. (f) Zhou, M.; Shang, L.; Li, B.; Huang, L.; Dong, S. Biosens. Bioelectron. 2008, 24 (3), 442−7. (5) Xia, Y.; Gates, B.; Sang Hyun, P. J. Lightwave Technol. 1999, 17 (11), 1956−1962. (6) (a) Kuroda, Y.; Kuroda, K. Angew. Chem., Int. Ed. 2010, 49 (39), 6993−7. (b) Kuroda, Y.; Sakamoto, Y.; Kuroda, K. J. Am. Chem. Soc. 2012, 134 (20), 8684−8692. (7) (a) Lin, Y. S.; Lu, C. Y.; Hung, Y.; Mou, C. Y. ChemPhysChem 2009, 10 (15), 2628−32. (b) Yang, S. M.; Coombs, N.; Ozin, G. A. Adv. Mater. 2000, 12 (24), 1940−1944. (c) Nakamura, T.; Yamada, H.; Yamada, Y.; Gurtanyel, A.; Hartmann, S.; Husing, N.; Yano, K. Langmuir 2010, 26 (3), 2002−7. (d) Trofimova, E. Y.; Kurdyukov, D. 2933
dx.doi.org/10.1021/cm500619p | Chem. Mater. 2014, 26, 2927−2933
