Preparation of Mesoporous Submicrometer Silica Capsules via an

Apr 10, 2012 - Mesoporous silica capsules with submicrometer sizes were successfully prepared via the interfacial hydrolysis and condensation reaction...
0 downloads 0 Views 3MB Size
Download PDF
Article pubs.acs.org/Langmuir

Preparation of Mesoporous Submicrometer Silica Capsules via an Interfacial Sol−Gel Process in Inverse Miniemulsion Zhihai Cao,*,† Lizi Dong,† Li Li,† Yue Shang,‡ Dongming Qi,§ Qun Lv,† Guorong Shan,‡ Ulrich Ziener,∥ and Katharina Landfester⊥ †

College of Materials, Chemistry and Chemical Engineering, Hangzhou Normal University, Xuelin Street 16, 310036 Hangzhou, China ‡ Department of Chemical Engineering, State Key Laboratory of Chemical Engineering, Zhejiang University, Hangzhou, Zhejiang 310027, China § Key Laboratory of Advanced Textile Materials and Manufacturing Technology, Ministry of Education, Zhejiang Sci-Tech University, Hangzhou, 310018, China ∥ Institute of Organic Chemistry III − Macromolecular Chemistry and Organic Materials, University of Ulm, Albert-Einstein-Allee 11, 89081 Ulm, Germany ⊥ Max Planck Institute for Polymer Research, Ackermannweg 10, 55128 Mainz, Germany S Supporting Information *

ABSTRACT: Mesoporous silica capsules with submicrometer sizes were successfully prepared via the interfacial hydrolysis and condensation reactions of tetraethoxysilane (TEOS) in inverse miniemulsion by using hydrophilic liquid droplets as template. The inverse miniemulsions containing pH-controlled hydrophilic droplets were first prepared via sonication by using poly(ethylene-co-butylene)-b-poly(ethylene oxide) (P(E/B)PEO) or SPAN 80 as surfactant. TEOS was directly introduced to the continuous phase of an inverse miniemulsion. The silica shell was formed by the deposition of silica on the surface of droplets. The formation of capsule morphology was confirmed by transmission electron microscopy (TEM) and field emission scanning electron microscopy (FESEM). The mesoporous structure was verified by nitrogen sorption measurements. The specific surface area could be tuned by the variation of the amount of cetyltrimethylammonium bromide (CTAB) and TEOS, and the pore size by the amount of CTAB. The influences of synthetic parameters on the particle size and morphology were investigated in terms of the amount of CTAB, pH value in the droplets, TEOS amount, surfactant amount, and type of solvent with low polarity. A formation mechanism of silica capsules was proposed.

■

simultaneously.8 Koh et al. precisely synthesized silica nanocapsules with a silanol-functionalized micelle template.9 Taking advantage of the easily tunable size, easy removal, as well as the possibility to directly incorporate functional cargos, soft templates, both the oily and aqueous droplets, have been intensively applied to prepare capsules or hollow particles in recent years. Schacht et al. prepared oil-filled mesoporous silica microcapsules by using the oil−water interface as template in a direct emulsion.10 Poulin et al. prepared mesoporous silica microcapsules with a homothetic morphology in inverse emulsion by mixing an acidic cetyltrimethylammonium bromide (CTAB) solution and a hydrophobic TEOS solution.11 In their case, the size of the silica capsules was in the range 2−50 μm due to the large size of the droplet template. A miniemulsion has been proven to be a very efficient technique

INTRODUCTION Much effort has been devoted to research on the elaboration of hollow particles or capsules due to their wide applications in the field of drug delivery systems, catalysis, biology, coatings, and so on.1−3 The versatile materials, such as polymeric, hybrid, and inorganic materials, have been used to form the shell of capsules.4−6 Among them, silica capsules are probably the most widely investigated inorganic capsules because of their high surface area, tunable pore size, high biocompatibility, and good optical property. Different synthetic techniques have been designed to prepare silica capsules, and most of them are based on template methods, including hard templates, micelle, or soft templates.7−16 Bourgeat-Lami et al. prepared core−shell particles via the sol−gel process of tetraethyl orthosilicate (TEOS) by using silanol-functionalized hybrid particles as template, and then calcined the core−shell particles to obtain hollow silica particles.7 Chen et al. fabricated hollow silica spheres via a one-step process by the formation of the silica shell and the dissolution of the core polystyrene particle © 2012 American Chemical Society

Received: February 5, 2012 Revised: April 1, 2012 Published: April 10, 2012 7023

dx.doi.org/10.1021/la300531b | Langmuir 2012, 28, 7023−7032

Langmuir

Article

Table 1. Formulations of All Experiments run

surfactant type

surfactant content (wt %)a

CTAB (wt %)b

low polarity solventc

TEOS (g)

pH adjuster/pHd (g)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

P(E/B)-PEO P(E/B)-PEO P(E/B)-PEO P(E/B)-PEO P(E/B)-PEO P(E/B)-PEO P(E/B)-PEO P(E/B)-PEO P(E/B)-PEO P(E/B)-PEO P(E/B)-PEO P(E/B)-PEO SPAN 80 P(E/B)-PEO P(E/B)-PEO

3.4 3.4 3.4 3.4 3.4 3.4 3.4 3.4 1.7 2.6 5.1 6.8 3.4 3.4 3.4

8.0 0.0 2.0 4.0 8.0 8.0 8.0 8.0 8.0 8.0 8.0 8.0 2.0 8.0 8.0

CH CH CH CH CH CH CH CH CH CH CH CH CH isopar M HD

4.0 1.0 1.0 1.0 1.0 1.0 1.0 2.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0

NH3·H2O/12.3 NH3·H2O/12.3 NH3·H2O/12.3 NH3·H2O/12.3 NH3·H2O/12.3 None/6.2 HCl/0.07 NH3·H2O/12.3 NH3·H2O/12.3 NH3·H2O/12.3 NH3·H2O/12.3 NH3·H2O/12.3 NH3·H2O/12.3 NH3·H2O/12.3 NH3·H2O/12.3

a On the basis of the mass of disperse phase. bOn the basis of the mass of aqueous solution of pH adjuster. cAmount of solvent with low polarity for all runs was 12.5 g. dMass of aqueous solution of pH adjuster was 1.25 g.

butylene)-b-poly(ethylene oxide) (P(E/B)-PEO) with the number average molecular weight of 6200 g·mol−1 determined by 1H NMR spectroscopy were synthesized according to the literature.34,35 The lengths of the hydrophobic (E/B) and hydrophilic block (EO block) are 4000 and 2200 g·mol−1, respectively, which results in an HLB of 7 for P(E/B)-PEO. Demineralized water with Milli-Q grade (resistivity: 18 MΩ) was used. Preparation of Inverse Miniemulsion. 0−0.1 g of CTAB was dissolved in 1.25 g of the pH-controlled water. The low HLB surfactant (P(E/B)-EPO in most cases; see Table 1) was dissolved in 12.5 g of CH, and the CH solution of low HLB surfactant was used as continuous phase. After 15 min pre-emulsification under strong magnetic stirring, the crude emulsion composed of the aqueous solution of CTAB and CH solution of low HLB surfactant was treated with ultrasound using a pulsed sequence (work 12 s, break 6 s) with a Scientz JY92-II DN sonifier at 67% duty (600 W) in an ice bath for 15 min to prepare an inverse miniemulsion. A pulsed mode of sonication was applied for all experiments to avoid overheating during sonication. The formulations are listed in Table 1. Preparation of Silica Capsules and Particles. After the preparation of inverse miniemulsion, TEOS was directly mixed with the continuous phase of inverse miniemulsion to prepare an inorganic shell on the surface of droplet templates. The whole mixture of the initial inverse miniemulsion and TEOS was placed into a closed glass reactor, and then kept in a preheated oil bath at 40 °C. The hydrolysis and condensation reactions of TEOS were run for 24 h with magnetic stirring at 500 rpm. To follow the evolution of the morphology, the samples were withdrawn at regular time intervals. Calcination of Silica Capsules. The dispersions with silica capsules were cleaned by five-cycle centrifugation and redispersion in CH. The cleaned and dried samples were heated to 550 °C at a heating rate of 2 °C·min−1, and then kept at 550 °C for 5 h in air to remove the organic components. Redispersion of Calcined Silica Capsules in Water. The calcined silica capsules were redispersed in water according to the following protocol. Twenty milligrams of calcined silica capsules was first mixed with 15 g of water with strong magnetic stirring for 30 min. The crude dispersion was treated with ultrasound using a pulsed sequence (work 12 s, break 6 s) at 50% duty (about 425 W) in an ice bath for 9 min. Characterization. Transmission Electron Microscopy (TEM). TEM measurements were performed on a JEOL JSM1230EXT20 microscope operating at 80 kV. One droplet of the original dispersion was diluted with 2 mL of CH, and then one droplet of the diluted sample was placed on a 400-mesh carbon-coated copper grid and dried at room temperature. The number-average size of capsules and number-average thickness

to prepare hollow particles and nanocapsules in the nanometer or submicrometer range by using droplets as the template.17−33 Polymeric and organic−inorganic hybrid nanocapsules have been prepared in both direct and inverse miniemulsion polymerizations by using hydrophobic or hydrophilic templates.17−29 Silica nanocapsules were also prepared via sol−gel process of a silica precursor in the direct miniemulsion by using octane, silicon oil, or hydrophobic solution of functional monomers as template.30−32 Schiller et al. prepared mesoporous silica particles and capsules via sol−gel process of a hydrophilic silica precursor in the dispersed phase of an inverse miniemulsion.33 To the best of our knowledge, the fabrication of submicrometer silica capsules based on the interfacial sol−gel process in an inverse miniemulsion system has not been reported yet. In the present paper, mesoporous submicrometer silica capsules with a good shell stability were successfully prepared via interfacial sol−gel process of TEOS in an inverse miniemulsion by using pH-controlled hydrophilic droplets as template. The size of silica capsules could be easily tuned from the range of submicrometer to micrometer by varying the CTAB amount. The shell of silica capsules was formed by the deposition of silica on the surface of droplet templates. The colloidal stability of the dispersion was mainly provided by the surfactants with a low hydrophilic−lipophilic balance (HLB) value (P(E/B)-PEO or SPAN 80), while CTAB was mainly used to form the capsule morphology. The silica capsules could be obtained under both acidic and basic conditions. The influences of synthetic parameters on the size and morphology were extensively investigated in terms of the CTAB amount, pH in the droplets, TEOS amount, surfactant amount and type, and type of solvent with low polarity.

■

EXPERIMENTAL SECTION

Materials. Tetraethoxysilane (TEOS, AR, Aladdin Chemistry Co. Ltd.), ammonia solution (25−28%, Jiangsu Qiangsheng Functional Chemistry Co. Ltd.), hydrochloric acid (HCl, 37%, Quzhou Juhua Chemical Reagent Co. Ltd.), cetyltrimethylammonium bromide (CTAB, 99.0%, Sinopharm Chemical Reagent Co. Ltd.), SPAN 80 (Shanghai Qiangshun Chemical Reagent Co. Ltd.), cyclohexane (CH, 99.5%, Shanghai Lingfeng Chemical Reagent Co. Ltd.), isopar M (a C12−C14 isoparaffinic mixture, Exxon Mobil), and hexadecane (HD, 99%, Acros Organics) were used as received. Poly(ethylene-co7024

dx.doi.org/10.1021/la300531b | Langmuir 2012, 28, 7023−7032

Langmuir

Article

Figure 1. TEM (a1,a2) and SEM (b1,b2) images of silica capsules prepared in inverse miniemulsion with different magnifications (see Table1, run1).

Figure 2. N2 adsorption−desorption isotherm (a) and pore size distribution (b) of the calcined silica capsules prepared in inverse miniemulsion (see Table 1, run1).

■

of shell were determined by counting 100 capsules in the TEM images. The standard deviations of capsule size and shell thickness were calculated by formula 1.

SD =

∑ (Xi − X )2 N−1

RESULTS AND DISCUSSION

Preparation of Mesoporous Submicrometer Silica Capsules by Using Hydrophilic Droplets as Template. In the present contribution, a crude inverse emulsion was first prepared by mixing a pH-controlled aqueous solution of CTAB and a CH solution of surfactant (P(E/B)-PEO or SPAN 80) by strong magnetic stirring, and then the inverse miniemulsion containing hydrophilic droplets was prepared by homogenizing the crude emulsion by sonication. The hydrophilic miniemulsion droplets not only contain one reactant (water) and catalyst (base or acid), but also work as liquid template to generate the capsule morphology. Due to the highly basic condition in the disperse phase, we did not attempt to introduce silica precursors directly into the disperse phase to avoid the premature condensation of silica precursors during the preparation of inverse miniemulsion. Instead, after the preparation of inverse miniemulsion, we added the silica precursor to the continuous phase of the inverse miniemulsions. By this strategy, hydrophilic silica precursor, like

(1)

In formula 1, X represents particle size or shell thickness; N represents the number of capsules. Field Emission Scanning Electron Microscopy (FESEM). FESEM measurements were performed on a field emission scanning electron microscopy (Ultra 55, Carl Zeiss SMT Pte Ltd.) by placing the powder of the dried samples on a conductive film. The accelerating voltage was 2 kV. Nitrogen Sorption Analysis. The calcined silica capsules were used to characterize the specific surface area and pore size by nitrogen sorption measurements at 77 K on a Quantachrome AUTOSORB-1-C automated gas sorption apparatus. The specific surface area was calculated by the Brunauer−Emmett−Teller (BET) equation. The pore size and pore size distribution were derived from the adsorption isotherm curves by using the Brunauer-Joyner-Halenda (BJH) method. 7025

dx.doi.org/10.1021/la300531b | Langmuir 2012, 28, 7023−7032

Langmuir

Article

Figure 3. TEM (a1, b1, c1, and d1) and SEM (a2, b2, c2, and d2) images of silica capsules synthesized in inverse miniemulsion with different CTAB amounts (a1 and a2, 0 wt % of CTAB; b1 and b2, 2.0 wt %; c1 and c2, 4.0 wt %; d1 and d2, 8.0 wt %; see Table 1, runs 2−5).

sodium silicate, cannot be used as a silica source, because they cannot dissolve into the hydrophobic continuous phase, and therefore cannot diffuse to the hydrophilic droplets to react with water. Tetraethoxysilane (TEOS), which is hydrophobic and completely miscible with CH, was used as the silica source to form an inorganic shell on the surface of liquid templates by a sol−gel process. TEOS diffused to the surface of the droplets, and underwent hydrolysis and condensation by contacting with water and catalyst (ammonia or hydrochloric acid) in the

hydrophilic droplets. The formation of the capsule morphology was confirmed by TEM; see Figure 1a1,a2. The formation mechanism of the silica capsules will be discussed in the following section. The number-average size of the capsules and number-average thickness of the shell determined by TEM are about 610 ± 120 nm and 63 ± 13 nm, respectively. The welldefined hollow spheres with mesoscale pores in shells could be easily seen in the TEM image with higher magnification (Figure 1a2). In addition, the shell of the capsules is also covered by a 7026

dx.doi.org/10.1021/la300531b | Langmuir 2012, 28, 7023−7032

Langmuir

Article

Figure 4. N2 adsorption−desoprtion isotherms (a) and pore size distribution (b) of the calcined silica capsules prepared in inverse miniemulsion with different CTAB amounts (see Table 1, runs 3−5).

large amount of tiny particles forming a raspberry-like structure; see Figure 1a2,b2. These tiny particles are believed to also be silica particles. Furthermore, the silica shell is stable enough to withstand the contraction of silica capsules during the drying process, and keeps the intact spherical structure, indicated by the absence of broken objects in the SEM images (Figure 1b1,b2). In addition, the calcined silica capsules could be well-redispersed in water, and retain the spherical morphology (SI Figure S1). No sediments were observed in the dispersion 5 h after preparation. The nitrogen adsorption−desorption isotherm and pore size distribution of the calcined silica capsules of run 1 are shown in Figure 2. The isotherm displays an IV isotherm, and the specific surface area and mean pore size of the calcined silica capsules are 120 m2·g−1 and 12 nm, respectively. The capillary condensation was observed in the isotherm of this sample, indicating the formation of mesoporous pores in the shell. In fact, the capillary condensation could be observed in the samples with different amounts of CTAB and TEOS (Figures 4a and 6a). This clearly shows that silica capsules with a mesoporous shell can be prepared by the present technique. Moreover, the specific surface area can be tuned by the variation of the amount of TEOS and CTAB, and the pore size by the variation of the CTAB amount (see below). Variables Influencing the Particle Properties. CTAB Amount. A strong interaction between the positively charged surfactant (S+) and negative inorganic silica species (I−) or the intermolecular interactions among the ion pairs (S+I−) can be formed under basic conditions. When the pH is below the isoelectric point of silica (about 236), the interaction between the cationic silica species and halide−cationic surfactant (S+X−I+) is mediated by the halide ions, which form hydrogen bonds with protonated silanols like SI(OH2)+.10,37 CTAB, as a commercial cationic surfactant, is frequently used as template to prepare mesoporous silica capsules or particles.10,11 As a surface active agent, CTAB preferentially distributes on the interface between the disperse and continuous phases. Therefore, CTAB was used as the morphology-directing agent to promote the deposition of silica species on the surface of droplets to form the capsule morphology in the present case. In order to verify the function of CTAB on the formation of capsule morphology, the system without adding CTAB (run 2) was performed for comparison. As expected, only solid silica particles were formed in this system (Figure 3a1,a2). This means that the silica species will diffuse into the interior of droplets due to the increase of their hydrophilicity with the

hydrolysis of TEOS in the system without the interaction between CTAB and silica species. It should be noted that hydrogen bonds might be formed between the silica or other inorganic species and the poly(ethylene oxide) block of P(E/ B)-PEO.38,39 However, obviously, this interaction is not strong enough to confine the deposition of silica on the surface of droplets. By introducing 2.0 wt % of CTAB relative to the aqueous solution of pH adjuster into the disperse phase, the morphology of particles successfully transformed from solid particle to capsule, as shown in Figure 3b1. Some broken particles in Figure 3b2 further confirmed the formation of the capsule morphology. The capsule morphology of particles always dominates in the products with adding different CTAB amounts (Figure 3c1,d1). As mentioned before, CTAB is a surface active agent, and partially contributes to the emulsification of the disperse phase. With the increase of the CTAB amount to 8.0 wt %, the size of silica capsules decreases to about 600 nm, much smaller than that of the silica capsules with 2.0 wt % and 4.0 wt % of CTAB (about 1.5 μm). Compared to the silica capsules with 2.0 wt % and 4.0 wt % of CTAB, more broken silica capsules appeared in the product with 8.0 wt % of CTAB (Figure 3b2,c2,d2). The breakage of silica capsules might take place in the drying process or the SEM sample preparation, but it clearly indicates that the shell stability of silica capsules is weak in the product with 8.0 wt % of CTAB. This could be ascribed to the increase of surface area of droplet templates due to the reduction in the template size, which leads to the decrease in silica amount in the shell of capsules. This analysis is confirmed by the fact that the shell stability can be significantly enhanced by using more TEOS. For example, in run 1, only intact silica capsules could be seen in the SEM images (Figure 1b1,b2). It should be pointed out that the amount of TEOS used in this series was 1 g. Closely observing the shell of silica capsules under high magnification (SI Figure S2) shows that the shell of silica capsules with 2.0 wt % and 4.0 wt % of CTAB is relatively smoother than that of the capsules with 8.0 wt % of CTAB. This means that the grains are more loosely packed with the increase of CTAB amount. This is confirmed by the fact that the pore size and specific surface area increase with the increase of CTAB amount. As shown in Figure 4, the mean pore size was about 3.5 nm, and the pore size distribution was uniform for the product with 2.0 wt % of CTAB. With the increase of CTAB amount to 4.0 wt %, a shoulder peak appeared on the branch of large size. The mean pore size from the peak of the small pores was about 3.2 nm, and the mean pore size of the 7027

dx.doi.org/10.1021/la300531b | Langmuir 2012, 28, 7023−7032

Langmuir

Article

Figure 5. TEM images of silica capsules synthesized in inverse miniemulsion with different pH conditions (a, pH 6.20; b, pH 0.07; see Table 1, runs 6−7).

shoulder peak was about 4.9 nm. The mean pore size further increased to about 12 nm when the CTAB amount increased to 8.0 wt %. In addition, the specific surface area increased from 210 m2·g−1 to 350 m2·g−1 and then to 540 m2·g−1 with the increase of CTAB amount from 2 wt % to 4 wt %, and then to 8 wt %. Type of pH Buffer. The hydrolysis and condensation rate of TEOS can be accelerated under acidic or basic condition. As shown in Figure 5a, the copper grids are covered by some irregular objects, which look like a film. This clearly indicates that the silica capsules did not form even after 24 h under neutral condition. As mentioned before, the silica capsules could be obtained by using basic droplets (pH 12.3) as template (Figure 3d1,d2). By decreasing the pH to 0.07, the silica capsules were successfully formed again (Figure 5b). Compared to the silica capsules synthesized under basic condition, the capsules synthesized under acidic condition were relatively soft and easy to deform. It should be pointed out that the pH value of the acidic solution was much lower than the isoelectric point of silica. It means the silica particles are positively charged under the acidic condition. The interaction between the silica species and CTAB (S+X−I+) was mainly governed by the halide ions. In addition, the reaction mechanisms of TEOS under acidic and basic condition are different. Under basic condition, TEOS is inclined to condense completely, and therefore, precipitates readily to form colloidal particles which establish the mesoporous shell via coassembly with CTAB through electrostatic interactions. Under acidic conditions, some silica oligomers are more probably formed due to the incomplete condensation of TEOS. These oligomers with unreacted silanols cannot precipitate easily, because they can partially dissolve in the ethanol/water solution formed by hydrolysis. With the increase of cross-linking extent, the silica oligomers will finally precipitate to form a silica shell under acidic conditions. The different interactions between silica species and CTAB, and the different reaction mechanism of TEOS, might lead to the differences in the pore structure and surface area of silica capsules synthesized under highly acidic condition from those synthesized under basic condition. The thorough investigations on the preparation of silica capsules under highly acidic conditions are ongoing, and more detailed information will be presented in a forthcoming paper. TEOS Amount. In principle, the size of droplets in the initial inverse miniemulsions does not depend on the TEOS amount, and therefore, the thickness of silica capsules could be possibly tuned by changing the TEOS amount. Surprisingly, the thickness of silica capsules only slightly increased with the

increase of TEOS amount; see Table 2. By comparing the morphology of particles with different amounts of TEOS Table 2. Relationship between the TEOS Amount and the Particles Size, Shell Thickness, Specific Surface Area, and Pore Size run

TEOS amount (g)

particle size (nm)

shell thickness (nm)

specific surface area (m2·g−1)

pore size (nm)

5 8 1

1 2 4

600 ± 140 619 ± 160 610 ± 120

44 ± 14 47 ± 12 63 ± 13

540 360 120

12 10 12

(Figure 1, 3d1,d2, and SI Figure S3), we could find that the shell stability of capsules increased with the increase of the TEOS amount. A large amount of broken silica capsules could be found in the product with 1 g of TEOS (Figure 3d1,d2), while the broken silica capsules were rarely seen in the products with 2 and 4 g of TEOS (Figure 1 and SI Figure S3). The silica capsules with the best-defined spherical morphology were produced by using 4 g of TEOS. In addition, we did not find free silica particles in the products with higher amount of TEOS. On the basis of these results, we speculate that a more compact and dense layer of silica should be built on the droplet templates in the systems with a high amount of TEOS, and therefore the resulting silica capsules show higher stability and can retain their particle morphology under the dried state. The formation of a more compact silica shell by using a higher TEOS amount is also confirmed by the reduction of specific surface area with the increase of the TEOS amount (Table 2). The products with different TEOS amount show a monomodal pore size distribution (Figure 6b), and the mean pore sizes of the silica capsules do not show a clear dependence on the amount of TEOS (see Table 2). Surfactant Amount and Type. P(E/B)-PEO has been proven to be a very efficient surfactant to stabilize inverse miniemulsions, due to its bulk molecular size and low HLB.33,40,41 The amount of P(E/B)-PEO required to prepare a stable miniemulsion could be even lower than 1 wt % relative to the disperse phase in some cases.33 Therefore, P(E/B)-PEO was used as surfactant to stabilize most of the initial inverse miniemulsions in this contribution. It was reported that P(E/ B)-PEO could work as a structure-directing agent to prepare highly organized mesoporous anatase films38 and porous silica by nanocasting.35 In inverse miniemulsion, P(E/B)-PEO mainly distributed on the interface between the oily and aqueous phases due to the large interfacial area of droplets 7028

dx.doi.org/10.1021/la300531b | Langmuir 2012, 28, 7023−7032

Langmuir

Article

Figure 6. N2 adsorption−desoprtion isotherms (a) and pore size distribution (b) of the calcined silica capsules prepared in inverse miniemulsion with different TEOS amounts (see Table 1, runs 1, 5, and 8).

Figure 7. TEM images of silica capsules synthesized in inverse miniemulsion with different surfactant amounts (a, 1.7 wt % of P(E/B)-PEO; b, 2.6 wt %; c, 5.1 wt %; d, 6.8 wt %; see Table 1, runs 9−12).

instead of the formation of polymer micelles by self-assembly. Schiller et al. reported that no obvious structure-directing effects were observed in their research on the preparation of silica particles in inverse miniemulsion.33 In the present case, the main function of P(E/B)-PEO is to work as surfactant to stabilize the droplets in inverse miniemulsion. However, due to the deposition of silica species from the outer surface of droplets, the presence of P(E/B)-PEO on the surface of droplets might influence the formation of pores to some extent in our case. The formation of the silica shell seems to be disturbed by the presence of P(E/B)-PEO, with the increase of the surfactant amount. A large amount of small solid silica particles appeared in the product with 6.8 wt % of P(E/B)-PEO to the aqueous disperse phase (Figure 7d). This might be due to the denser layer of P(E/B)-PEO on the surface of droplet templates in this system, which interferes with the interaction between silica and CTAB. Roughly estimated by TEM, the size of silica capsules decreased with the increase of surfactant

amount. This could be reasonably ascribed to the smaller droplets in the initial miniemulsion with higher surfactant amount. SPAN 80 could also be used as surfactant to stabilize the droplets in inverse miniemulsion. In addition, compared to the silica capsules stabilized by P(E/B)-PEO, the size of silica capsules was significantly reduced to about 200 nm, determined by TEM (SI Figure S4). The silica capsules were inclined to deform and a part of the silica capsules was incomplete, most probably due to the thin shell of the capsules, similar to run 5. In addition, the adsorption of CTAB on the interface may be limited by the more unfavorable curvature with the decrease of droplet size. This might influence the deposition of silica on the surfaces of the droplets, leading to formation of an incomplete shell. Type of Solvent with Low Polarity. The size of the silica capsules with isopar M as low polarity solvent was much larger than that of the system with CH (run 5), and in addition, the 7029

dx.doi.org/10.1021/la300531b | Langmuir 2012, 28, 7023−7032

Langmuir

Article

Figure 8. Morphological evolution of silica capsules during the reaction (see Table 1, run 1).

Scheme 1. Schematic Representation of the Formation of Submicrometer Silica Capsules in Inverse Miniemulsion

stability might be caused by using an unsuitable surfactant if isopar M or HD is employed as low polarity solvent. The shell of the silica capsules is composed of silica particles with a size of about 90 nm, much bigger than the silica particles in the system with CH (less than 20 nm). In addition, some free silica particles could also be observed in the TEM image (see SI Figure S5). Mechanism of the Formation of Silica Capsules. As shown in Figure 8a, some soft capsules with a thin shell have already formed in 15 min. As the reaction proceeded, the amount of soft capsules increased significantly, and the shell thickness increased gradually, indicated by the increase of shell contrast. The silica capsules reacted for less than 2 h deformed more or less to elliptical shapes in the dried state. This could be ascribed to the small thickness of the silica shell and the relatively low cross-linking extent in the shell. With longer reaction time, spherical capsules dominated in the products; see

colloidal stability of this system is relatively worse (SI Figure S5a). This observation is consistent with our previous results on the inverse miniemulsion polymerization of 2-hydroxy methacrylate.41 In that case, we found that the coagulum amount produced during the polymerization increased from about 3 wt % relative to monomer for the systems with CH as low polarity solvent to about 20 wt % for the systems with isopar M. We obtained similar results in the system with HD as low polarity solvent (SI Figure S5b). The large size and poor colloidal stability could be ascribed to the relatively higher hydrophobicity of isopar M and HD than that of CH, which may lead to a higher interfacial tension between the disperse and continuous phases in the system with isopar M or HD as low polarity solvent.41 According to our previous finding,41 the particle size and colloidal stability also strongly depend on the combination of polar solvent, low polarity solvent, and surfactant, and therefore, the large size and poor colloidal 7030

dx.doi.org/10.1021/la300531b | Langmuir 2012, 28, 7023−7032

Langmuir

Article

ammonia or to 0.07 by using hydrochloric acid, the rate of hydrolysis and condensation reactions of TEOS was fast enough to build a silica shell on the surface of the droplet templates in 24 h. CTAB as the morphology-directing agent is crucial for the formation of capsule morphology, and only solid particles could be formed in the system without CTAB. The size of the silica capsules could be reduced by the increase of the CTAB amount. In addition, the pore size distribution and specific surface area could also be tuned by the variation of CTAB amount. The amount of TEOS played a very important role in building a stable silica shell on the droplet templates. Most of the silica capsules were broken in the SEM image in the product with 1 g of TEOS. This situation was improved by increasing the TEOS amount to 4 g, indicated by the fact that most of the silica capsules became intact in the SEM image. In addition, by varying the TEOS amount, the specific surface area of the silica capsules could be tuned from 120 to 540 m2·g−1. The formation of the silica shell would interfere with the increase of surfactant amount. Besides P(E/B)-PEO, silica capsules could also be obtained by using SPAN 80 as surfactant. The control of the synthesis of silica capsules became more difficult by using a more hydrophobic low-polarity solvent, isopar M or HD, than CH. By following the morphological evolution during the reaction, we found that a soft silica shell has already formed in 15 min, and with prolonging the reaction time, a thicker and more cross-linked shell was built on the surface of the droplet templates. The presented technique of this paper has potential application to directly incorporate hydrophilic functional cargos.

Figure 8f−i. In Schiller’s case, the formation of capsules was realized by the sol−gel process of silica precursor in the disperse phase.33 There, the particle morphology evolved from particles (after about 15 min) to a mixture of porous particles and capsules (after 2 h), finally to capsules (after 1 day). According to the results in Figure 8, we consider that the formation mechanism of silica capsules in our case is completely different from that in Schiller’s case. The proposed mechanism of formation of silica capsules in our case is shown in Scheme 1. TEOS is dissolved in the oil phase, which prevents hydrolysis and condensation in the oil phase due to the absence of water and catalyst. Only the TEOS molecules, which reach the surface of the droplets by diffusion, can undergo hydrolysis of the ethoxysilyl groups and subsequently condense to form an inorganic cross-linked network. Under basic conditions, the formation of a silica shell on the droplets is dominated by strong interaction between the positively charged surfactant (CTAB in the present case, S+) and negative inorganic silica species (I−), and the intermolecular interactions among the ion pairs (S+I−).10 In the direct system, this strong interaction between surfactant and silica particles may probably abstract the surfactant from the droplets, leading to colloidal instability of the oil droplets. However, in our case, CTAB is mainly used to direct the formation of silica shell, and the colloidal stability of the inverse miniemulsion is provided by the surfactant with a low HLB value, such as the amphiphilic block copolymer surfactant P(E/B)-PEO, or SPAN80. The formation of the silica shell on the droplets starts with the hydrolysis and condensation reactions of TEOS at the beginning, and therefore, the particles with a capsule morphology can be observed in the sample with 15 min reaction time. Prolonging the reaction time mainly contributes to building a thicker shell and improving the condensation extent in the shell. In addition, the rate of formation of silica shell also depends strongly on the amount of TEOS, and increases with the increase of the TEOS amount (Figure 8 and SI Figure S6). For the system with 4 g of TEOS (run 1), a relatively stable shell has formed after 2 h, while for the system with 1 g of TEOS (run 5), some deformed capsules could still be seen in the sample reacted for 8 h. In order to ensure the formation of a stable shell, we run the reaction for 24 h in all cases. Although miniemulsion is a kind of kinetically stable system, normally the droplets can remain stable for at least several hours.42 In our case, the reaction temperature is relatively low, 40 °C, and a thin silica shell has already formed in 15 min. The formation of a silica shell could also improve the stability of the disperse phase. Therefore, we believe that the droplet identity could be preserved in our case. It should be pointed out that we also observe that the silica capsules aggregate to form some micro-objects in the dispersion. This might be due to the interference of the colloidal protection from P(E/B)-PEO and SPAN 80 by the formation of silica on the surface of droplet templates.

■

ASSOCIATED CONTENT

S Supporting Information *

Additional TEM images. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The financial support from National Natural Scientific Foundation of China (NNSFC) project (51003023), the Hangzhou Normal University high-level talents start-up fund (2011QDL04), and State Key Laboratory of Chemical Engineering (SKL-ChE-12D04) are gratefully acknowledged. We greatly thank Dr. Yifeng Shi for the helpful discussion on the N2 adsorption data.

■

■

CONCLUSION Mesoporous silica capsules with a hydrophilic core were prepared via an interfacial sol−gel process of a silica precursor in inverse miniemulsion by using droplets as soft template. Due to the good miscibility of TEOS with the hydrophobic continuous phase, it was directly introduced into the inverse miniemulsions. The formation of capsule morphology was confirmed by TEM and FESEM. The formation of a mesoporous structure of the silica shell was verified by nitrogen sorption measurements. By tuning the pH to 12.3 by using

REFERENCES

(1) Landfester, K. Miniemulsion Polymerization and the Structure of Polymer and Hybrid Nanoparticles. Angew. Chem., Int. Ed. 2009, 48, 4488−4507. (2) Zhu, A.; Pan, Y.; Dai, S.; Li, F.; Shen, J. Preparation of NMaleoylchitosan Nanocapsules for Loading and Sustained Release of Felodipine. Biomacromolecules 2009, 10, 1997−2002. (3) Torchilin, V. Multifunctional and Stimuli-Sensitive Pharmaceutical Nanocarriers. Eur. J. Pharm. Biopharm. 2009, 71, 431−444. (4) Xu, X.; Asher, S. A. Synthesis and Utilization of Monodisperse Hollow Polymeric Particles in Photonic Crystals. J. Am. Chem. Soc. 2004, 126, 7940−7945.

7031

dx.doi.org/10.1021/la300531b | Langmuir 2012, 28, 7023−7032

Langmuir

Article

Prediction, Synthesis, and Characterization. Macromolecules 2003, 36, 8621−8629. (25) Luo, Y. W.; Gu, H. Y. A General Strategy for NanoEncapsulation via Interfacially Confined Living/Controlled Radical Miniemulsion Polymerization. Macromol. Rapid Commun. 2006, 27, 21−25. (26) Crespy, D.; Stark, M.; Hoffmann-Richter, C.; Ziener, U.; Landfester, K. Polymeric Nanoreactors for Hydrophilic Reagents Synthesized by Interfacial Polycondensation on Miniemulsion Droplets. Macromolecules 2007, 40, 3122−3135. (27) Cao, Z. H.; Ziener, U.; Landfester, K. Synthesis of Narrowly Size-Distributed Thermosensitive Poly(N-isopropylacrylamide) Nanocapsules in Inverse Miniemulsion. Macromolecules 2010, 43, 6353− 6360. (28) Rosenbauer, E.-M.; Landfester, K.; Musyanovych, A. SurfaceActive Monomer as a Stabilizer for Polyurea Nanocapsules Synthesized via Interfacial Polyaddition in Inverse Miniemulsion. Langmuir 2009, 25, 12084−12091. (29) Lu, F.; Luo, Y.; Li, B.; Zhao, Q.; Schork, F. J. Synthesis of Thermo-Sensitive Nanocapsules via Inverse Miniemulsion Polymerization Using a PEORAFT Agent. Macromolecules 2010, 43, 568− 571. (30) Peng, B.; Chen, M.; Zhou, S.; Wu, L.; Ma, X. Fabrication of Hollow Silica Spheres Using Droplet Templates Derived from a Miniemulsion Technique. J. Colloid Interface Sci. 2008, 321, 67−73. (31) Pang, H.; Zhou, S.; Wu, L.; Chen, M.; Gu, G. Fabrication of Silicone Oil Microcapsules with Silica Shell by Miniemulsion Method. Colloids Surf., A: Physicochem. Eng. Aspects 2010, 364, 42−48. (32) Fickert, J.; Rupper, P.; Graf, R.; Landfester, K.; Crespy, D. Design and Characterization of Functionalized Silica Nanocontainers for Self-Healing Materials. J. Mater. Chem. 2012, 22, 2286−2291. (33) Schiller, R.; Weiss, C. K.; Geserick, J.; Hüsing, N.; Landfester, K. Synthesis of Mesoporous Silica Particles and Capsules by Miniemulsion Technique. Chem. Mater. 2009, 21, 5088−5098. (34) Schlaad, H.; Kukula, H.; Runloff, J.; Below, I. Synthesis of α,ωHeterobifunctional Poly(ethylene glycol)s by Metal-Free Anionic Ring-Opening Polymerization. Macromolecules 2001, 34, 4302−4304. (35) Thomas, A.; Schlaad, H.; Smarsly, B.; Antonietti, M. Replication of Lyotropic Block Copolymer Mesophases into Porous Silica by Nanocasting: Learning about Finer Details of Polymer Self-Assembly. Langmuir 2003, 19, 4455−4459. (36) Park, G. A. The Isoelectric Points of Solid Oxides, Solid Hydroxides, and Aqueous Hydroxo Complex Systems. Chem. Rev. 1965, 65, 177−198. (37) Huo, Q.; Margolese, D. I.; Clesla, U.; Peng, P.; Fier, T. E.; Sieger, P.; Leon, R.; Petroff, P. M.; Schüth, F.; Stucky, G. D. Generilized Synthesis of Periodic Surfactant/Inorganic Composite Materials. Nature 1994, 368, 317−321. (38) Smarsly, B.; Frosso, D.; Brezesinski, T.; Pinna, N.; Boissière, C.; Antonietti, M.; Sanchez, C. Highly Crystalline Cubic Mesoporous TiO2 with 10-nm Pore Diameter Made with a New Block Copolymer Template. Chem. Mater. 2004, 16, 2948−2952. (39) Rossmanith, R.; Weiss, C. K.; Geserick, J.; Hüsing, N.; Hörmann, U.; Kaiser, U.; Landfester, K. Porous Anatase Nanoparticles with High Specific Surface Area Prepared by Miniemulsion Technique. Chem. Mater. 2008, 20, 5768−5780. (40) Landfester, K.; Willert, M.; Antonietti, M. Preparation of Polymer Particles in Nonaqueous Direct and Inverse Miniemulsions. Macromolecules 2000, 33, 2370−2376. (41) Cao, Z. H.; Wang, Z.; Herrmann, C.; Ziener, U.; Landfester, K. Narrowly size-distributed cobalt salt containing poly(2-hydroxyethyl methacrylate) particles by inverse miniemulsion. Langmuir 2010, 26, 7054−7061. (42) Schork, F. J.; Luo, Y. W.; Smulders, W.; Russum, J. P.; Butté, A.; Fontenot, K. Miniemulsion Polymerization. Adv. Polym. Sci. 2005, 175, 129−255.

(5) Lou, X. W.; Archer, L. A. A General Route to Nonspherical Anatase TiO2 Hollow Colloids and Magnetic Multifunctional Particles. Adv. Mater. 2008, 20, 1853−1858. (6) Tan, B.; Vyas, S. M.; Lehmler, H.-J.; Knutson, B. L.; Rankin, S. E. Synthesis of Inorganic and Organic−Inorganic Hybrid Hollow Particles Using a Cationic Surfactant with a Partially Fluorinated Tail. Adv. Funct. Mater. 2007, 17, 2500−2508. (7) Tissot, I.; Reymond, J. P.; Lefebvre, F.; Bourgeat-Lami, E. SiOHFunctionalized Polystyrene Latexes. A Step toward the Synthesis of Hollow Silica Nanoparticles. Chem. Mater. 2002, 14, 1325−1331. (8) Chen, M.; Wu, L.; Zhou, S.; You, B. A Method for the Fabrication of Monodisperse Hollow Silica Spheres. Adv. Mater. 2006, 18, 801−806. (9) Koh, K.; Ohno, K.; Tsujii, Y.; Fukuda, T. Precision Synthesis of Organic/Inorganic Hybrid Nanocapsules with a Silanol-Functionalized Micelle Template. Angew. Chem., Int. Ed. 2003, 42, 4194−4197. (10) Schacht, S.; Huo, Q.; Voigt-Martin, G.; Stucky, G. D.; Schüth, F. Oil-Water Interface Templating of Mesoporous Macroscale Structures. Science 1996, 273, 768−771. (11) Fornasieri, G.; Badaire, S.; Backov, R.; Mondain-monval, O.; Zakri, C.; Poulin, P. Mesoporous and Homothetic Silica Capsules in Reverse-Emulsion Microreactors. Adv. Mater. 2004, 16, 1094−1097. (12) Khanal, A.; Inoue, Y.; Yada, M.; Nakashima, K. Synthesis of Silica Hollow Nanoparticles Templated by Polymeric Micelle with Core−Shell−Corona Structure. J. Am. Chem. Soc. 2007, 129, 1534− 1535. (13) Blas, H.; Save, M.; Pasetto, P.; Boissiere, C.; Sanchez, C.; Charleux, B. Elaboration of Monodisperse Spherical Hollow Particles with Ordered Mesoporous Silica Shells via Dual Latex/Surfactant Templating: Radial Orientation of Mesopore Channels. Langmuir 2008, 24, 13132−13137. (14) Wang, J.-G.; Li, F.; Zhou, H.-J.; Sun, P.-C.; Ding, D.-T.; Chen, T.-H. Silica Hollow Spheres with Ordered and Radially Oriented Amino-Functionalized Mesochannels. Chem. Mater. 2009, 21, 612− 620. (15) Fujiwara, M.; Shiokawa, K.; Sakakura, I.; Nakahara, Y. Silica Hollow Spheres with Nano-Macroholes Like Diatomaceous Earth. Nano Lett. 2006, 6, 2925−2928. (16) Liu, J.; Hartono, S. B.; Jin, Y. G.; Li, Z.; Lu, G. Q.; Qiao, S. Z. A Facile Vesicle Template Route to Multi-Shelled Mesoporous Silica Hollow Nanospheres. J. Mater. Chem. 2010, 20, 4595−4601. (17) Tiarks, F.; Landfester, K.; Antonietti, M. Preparation of Polymeric Nanocapsules by Miniemulsion Polymerization. Langmuir 2001, 17, 908−918. (18) Ni, K. F.; Shan, G. R.; Weng, Z. X. Synthesis of Hybrid Nanocapsules by Miniemulsion (Co)polymerization of Styrene and γMethacryloxypropyltrimethoxysilane. Macromolecules 2006, 39, 2529− 2535. (19) Cao, Z.; Shan, G. R. Synthesis of Polymeric Nanocapsules with a Crosslinked Shell through Interfacial Miniemulsion Polymerization. J. Polym. Sci., Part A: Polym. Chem. 2009, 47, 1522−1534. (20) Cao, Z. H.; Shan, G. R.; Sheibat-Othman, N.; Putaux, J. L.; Bourgeat-Lami, E. Synthesis of Oily Core-Hybrid Shell Nanocapsules through Interfacial Free Radical Copolymerization in Miniemulsion: Droplet Formation and Nucleation. J. Polym. Sci., Part A: Polym. Chem. 2010, 48, 593−603. (21) Cao, Z.; Schrade, A.; Landfester, K.; Ziener, U. Synthesis of Raspberry-Like Organic−Inorganic Hybrid Nanocapsules via Pickering Miniemulsion Polymerization: Colloidal Stability and Morphology. J. Polym. Sci., Part A: Polym. Chem. 2011, 49, 2382−2394. (22) Schrade, A.; Cao, Z.; Landfester, K.; Ziener, U. Preparation of Raspberry-like Nanocapsules by the Combination of Pickering Emulsification and Solvent Displacement Technique. Langmuir 2011, 27, 6689−6700. (23) Li, W.; Matyjaszewski, K.; Albrecht, K.; Möller, M. Reactive Surfactants for Polymeric Nanocapsules via Interfacially Confined Miniemulsion ATRP. Macromolecules 2009, 42, 8228−8233. (24) van Zyl, A. J. P.; Sanderson, R. D.; Wet-Roos, D.; Klumperman, B. Core/Shell Particles Containing Liquid Cores: Morphology 7032

dx.doi.org/10.1021/la300531b | Langmuir 2012, 28, 7023−7032