Synthesis of highly-ordered mesoporous silica particles using mixed cationic and anionic surfactants as templates

Synthesis of highly-ordered mesoporous silica particles using mixed cationic and anionic surfactants as templates

Journal of Colloid and Interface Science 312 (2007) 42–46 www.elsevier.com/locate/jcis Synthesis of highly-ordered mesoporous silica particles using ...

373KB Sizes 2 Downloads 22 Views

Journal of Colloid and Interface Science 312 (2007) 42–46 www.elsevier.com/locate/jcis

Synthesis of highly-ordered mesoporous silica particles using mixed cationic and anionic surfactants as templates Takahiro Ohkubo a,∗ , Taku Ogura b , Hideki Sakai a,b , Masahiko Abe a,b a Institute of Colloid and Interface Science, Tokyo University of Science, 2641 Yamazaki, Noda, Chiba 278-8510, Japan b Faculty of Science and Technology, Tokyo University of Science, 2641 Yamazaki, Noda, Chiba 278-8510, Japan

Received 31 May 2006; accepted 12 February 2007 Available online 21 February 2007

Abstract We applied a molecular assembly formed in an aqueous surfactant mixture of cationic cetyltrimethylammonium bromide (CTAB) and anionic sodium octylsulfate (SOS) as templates of mesoporous silica materials. The hexagonal pore size can be controlled between 3.22 and 3.66 nm with the mixed surfactant system. In addition, we could observe the lamellar structure of the mixed surfactants with precursor molecules, which strongly shows the possibility of precise control of both the pore size and the structure of pores by changing the mixing ratio of surfactants. Moreover, use of the cationic surfactant having longer hydrophobic chain like stearyltrimethylammonium bromide (STAB) caused the increase in d100 space and shifted the point of phase transition from hexagonal phase to lamellar phase to lower concentration of SOS. © 2007 Elsevier Inc. All rights reserved. Keywords: Mesoporous silica; Mixed surfactant system; Templating method; X-ray diffraction

1. Introduction Porous materials have been widely used in catalytic reactions as well as in separation and purification processes. Remarkable progress in the research and development of nanoporous materials has been made in both fundamental and practical aspects since the publication on carbon nanotubes [1,2], MCMs [3,4], and FSMs [5]. In particular, single- or multicomponent metal-oxide mesoporous materials with controlled morphologies are indispensable for many catalytic reactions because they yield a good reaction rate stemming from a better diffusion velocity of adsorbed molecules and a highly dispersed state of catalytic components in mesopores. Recently, a variety of metal-oxide mesoporous solids with regular pore structures have been developed. These solids have been synthesized by a molecular template (MT) method, because the pore structure can easily be controlled by properly choosing the experimental conditions such as template concentration, precursor, pH, and temperature [6–11]. When we * Corresponding author. Fax: +81 4 7122 1442.

E-mail address: [email protected] (T. Ohkubo). 0021-9797/$ – see front matter © 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.jcis.2007.02.043

use surfactant as a MT source, a surfactant having a longer hydrophobic chain generally gives larger mesopores [12]. Although we can control the pore structure by use of different kinds of templating surfactants, it is difficult to obtain materials with desired pore sizes and shape because of the structural freedom of longer hydrophobic chains of surfactants. Meanwhile, physical properties of mixed cationic and anionic surfactants have attracted much attention [13–15]. Molecular assembly of the mixed surfactants can form a variety of structures such as cylindrical micelles, vesicles, and planar bilayers depending on the molar mixing ratio in the solution, because the Coulombic interaction between the hydrophilic parts of anionic and cationic surfactants causes the formation of quasi-double-chained composites, whose properties are greatly different from those of single-component systems. Recently, mixed surfactant systems such as cationic–cationic [12,16] or cationic–nonionic [17–19] mixtures were used as MT method to produce mesoporous materials. Especially, cationic–anionic surfactant systems were used to control the pore sizes of mesopores, precisely [20–22]. For instance, Chen et al. [23] used cetyltrimethylammonium bromide (CTAB) and sodium laurate (SL) as cationic and anionic surfactants, respectively. They re-

T. Ohkubo et al. / Journal of Colloid and Interface Science 312 (2007) 42–46

ported that SL could work as a polar organic cosolvent and/or cosurfactant. On the other hand, an aqueous mixture of cationic CTAB with relatively long alkyl chain (n = 16) and sodium octylsulfate (SOS) with short alkyl chain (n = 8) gives various kinds of molecular assemblies like worm-like micelle, vesicle, and lamellar liquid crystal at wider mixing composition [13,14] because the difference of alkyl chain length prevents the formation of insoluble salts between cationic and anionic surfactants. Therefore, if we prepare mesoporous silica using the CTAB/SOS system as structure directing agent, we can control the pore geometry and pore size more precisely by changing molar ratio of the two surfactants. In this paper, we report the characteristic properties of mesoporous silica materials synthesized by a MT method using CTAB/SOS and some other “catanionic” surfactant systems at different molar mixing ratios. 2. Experimental CTAB (Sigma-Aldrich Chemical Co.) and sodium octyl sulfate (SOS, Sigma-Aldrich Chemical Co.) were respectively used as the cationic and anionic surfactants for molecular template formation. In addition, stearyltrimethylammonium bromide (STAB, Sigma-Aldrich Chemical Co.) and myristyltrimethylammonium bromide (MTAB, Sigma-Aldrich Chemical Co.) were used as cationic surfactants to discuss the effect of chain length of hydrophobic parts to the formation of MTs. Tetraethyl orthosilicate (TEOS) was used as the silica precursor. Mixture of each cationic surfactant and SOS was dissolved in 25 ml of distilled water for injection (pH 5.6, Otsuka Pharm. Co.) with 25 ml of NaOH aqueous solution (pH 13) at the fixed total concentration of the mixed surfactants (total concentration = 60 mM). The compositions of the mixed surfactants are listed in Table 1. To the mixed surfactant solution was added 2.735 g of TEOS, and the mixture was stirred until TEOS was completely dissolved (over 24 h). The silica-mixed surfactant products were then filtered, washed with water, and dried at 393 K for 10 h. The silica particles were obtained from mixed particles after being calcined at 773 K for 6 h. Here, the silica-mixed surfactant products and mesoporous silica particles obtained are denoted by Si/MS-x and MPS-x, respectively. Here, x means the alphabetical sample order listed in Table 1. For instance, MPS-A means the mesoporous silica particles made from an aqueous mixed solution at CTAB:SOS = 10:0. No silica-mixed surfactant product was obtained with an aqueous solution at a molar ratio less than one, because the repulsive force between TEOS molecules and the surface of molecular assembly of anionic and cationic mixed surfactants prevents the ordered structure of TEOS molecules from being formed near the surface. Therefore, the characterization of both Si/MS-x and MPS-x was performed only for the samples synthesized using molar mixing ratios shown in Table 1. Powder XRD patters were obtained on a Rigaku RINT1100 diffractometer using CuKα radiation (40 kV, 25 mA). The images of pore structure and particles were obtained with a JEOL JEM-1200EX transmission electron microscope. Also, mesoporous structures for MPS-x were characterized by means

43

Table 1 Mixing ratios of anionic and cationic surfactants x

CTAB:SOS (molar ratio)

A B C D E F G H I J K

10:0 9.5:0.5 9:1 8.5:1.5 8:2 7.5:2.5 7:3 6.5:3.5 6:4 5.5:4.5 5:5

of N2 adsorption–desorption isotherm measurements at 77 K with an Quantachrome Autosorb-1-MP. Each sample for adsorption isotherm measurement was evacuated under 10−5 Torr at 425 K over 2 h. The adsorption/desorption data was analyzed using nonlocal density functional theory (NLDFT) approach to obtain the pore-size distribution (PSD) [24–26]. 3. Results and discussion Fig. 1 shows the powder XRD patterns of Si/MS-x samples. The pattern for Si/MS-A (CTAB:SOS = 10:0) was found to have four peaks at 2.30◦ , 3.96◦ , 4.60◦ , and 6.08◦ due to (100), (110), (200), and (210) reflections, respectively, indicating a highly ordered hexagonal structure. On the other hand, the patterns for samples between Si/MS-D and Si/MS-E clearly showed a tendency of structural change depending on the (100) reflection of hexagonal structure. Also, the patterns for samples from Si/MS-E to Si/MS-K can be referred to those for lamellar structure. Here, we can classify hexagonal or lamellar phases from peak positions of XRD; we can observe the main peak from (100) reflection and additionally (200) peak for lamellar phase. On the other hand, XRD patterns can show other peak positions such as (110) and (210) reflections for hexagonal phase. Table 2 summarizes the interlayer distances calculated from the (100) reflections of Si/MS-x samples. These results suggest that we can precisely control the interlayer distance of hexagonal and lamellar structures of Si/MS particles by increasing or decreasing the SOS mixing ratio in the precursor. These results also indicate that a drastic change of the pore structure can be observed only when the system of cationic and anionic mixed surfactants is used. Fig. 2 shows typical TEM images of Si/MS-B and Si/MS-F. These images indicated highly ordered structures of the mixed surfactant systems supported with TEOS. Although the lattice structures of these two samples are different from each other from the results of XRD, a highly ordered hexagonal or lamellar structure can be formed depending on the mixing ratio of CTAB and SOS. Hence, both the lattice structure and lattice constant of Si/MS-x can be defined by the mixing ratio of anionic and cationic surfactants and the pore structure of silica particles synthesized using Si/MS samples can be highly regulated by properly choosing the mixing ratio of the surfactants.

44

T. Ohkubo et al. / Journal of Colloid and Interface Science 312 (2007) 42–46

(a)

(b) Fig. 2. TEM images of Si/MS-x: (a) Si/MS-B (9.5:0.5) and (b) Si/MS-F (7.5:2.5).

Fig. 1. Powder XRD patterns of Si/MS-x samples.

Table 2 Peak positions (2θ ) and interlayer distances (d100 ) of Si/MS-x samples obtained from each (100) reflection Si/MS-x

2θ (◦ )

d100 (nm)

A (10:0) B (9.5:0.5) C (9:1) D (8.5:1.5) E (8:2) F (7.5:2.5) G (7:3) H (6.5:3.5) I (6:4) J (5.5:4.5) K (5:5)

2.30 2.22 2.14 2.10 2.44 2.26 2.16 1.96 1.92 1.88 1.84

3.84 3.98 4.13 4.25 3.62 3.91 4.09 4.51 4.60 4.69 4.80

We calcined each Si/MS-x sample to remove templating surfactant molecules and obtained silica particles with highly ordered mesopores. Fig. 3 shows typical TEM images of MPS-B and MPS-F synthesized from Si/MS-B and Si/MS-F, respectively. While a highly ordered pore structure for MPS-B was observed, no pore structure for MPS-F was found after calcination. These results can be supported by the powder XRD

patterns of MPS-x samples shown in Fig. 4. The pore sizes calculated from the (100) reflection of hexagonal lattice increased with increasing the ratio of SOS. Although we were able to obtain particles from Si/MS-F which is the precursor of MPS-F, the pore structure vanished after calcination. These TEM and XRD results strongly suggest that templating surfactant molecules support the voids of the lamellar structure formed between molecular assemblies of the mixed surfactants. Khushalani et al. reported that the hexagonal structure of siliceous mesoporous material is upheld after the calcination at 813 K [27]. However, generally speaking, the pore structure forming a lamellar lattice cannot be obtained if there are no pillars to keep the pore structure. Table 3 summarizes the analytical results of the powder XRD patterns shown in Fig. 4. The pore size in the hexagonal region (from MPS-A to MPS-D) gradually increased from 3.22 to 3.66 nm with increasing SOS mixing ratio. In addition, the pore structure of MPS-E could remain though Si/MS-E was referred to as a lamellar phase as mentioned above. Therefore, this suggests that Si/MS-E forms an intermediate state between hexagonal and lamellar phases, and the pore structure can be formed even after calcination to remove the templating molecules. Thus, we can precisely control the pore size of the hexagonal lattice for MPS samples by changing the mixing ratio of anionic and cationic surfactants.

T. Ohkubo et al. / Journal of Colloid and Interface Science 312 (2007) 42–46

45

(a)

(b) Fig. 3. TEM images of MPS-x: (a) MPS-B (9.5:0.5) and (b) MPS-F (7.5:2.5). Table 3 Peak positions (2θ ) and interlayer distances (d100 ) of MPS-x samples obtained from each (100) reflection MPS-x

2θ (◦ )

d100 (nm)

A (10:0) B (9.5:0.5) C (9:1) D (8.5:1.5) E (8:2)

2.30 2.22 2.14 2.10 2.44

3.84 3.98 4.13 4.25 3.62

F (7.5:2.5) ... K (5:5)

Fig. 4. Powder XRD patterns of MPS-x samples.

Not available

Fig. 5 shows N2 adsorption–desorption isotherm of MPSB at 77 K and PSD calculated by the isotherm is also shown in Fig. 6. We can observe no hysteresis loop in the isotherm, indicating that MPS-B sample has relatively smaller mesopores and/or micropores. Also, the peak position of the largest peak in PSD was 3.2 nm which is similar to the value of interlayer distance (d100 = 3.32 nm) calculated by the peak position of XRD. It has been reported that pore-wall thickness of mesoporous materials can be calculated from d100 and pore width obtained by XRD and gas adsorption measurements, respectively. However, we cannot obtain reasonable value from our data. This is because we synthesized the material from mixed surfactant systems in this study and the structure of pore wall of MPS-x was partially changed to distort the pore structure although XRD and TEM data strongly indicated hexagonal structure.

Fig. 5. Adsorption and desorption isotherm of N2 on MPS-B at 77 K: (") adsorption branch and (!) desorption branch.

Finally, we determined the effect of chain length of cationic surfactants by using MTAB and STAB while keeping anionic surfactant SOS. Fig. 7 shows the relationship between the d100 space of each composite material and the mixing molar ratio

46

T. Ohkubo et al. / Journal of Colloid and Interface Science 312 (2007) 42–46

mesoporous silica particles. The pore structure of the silica particles synthesized was changed with increasing or decreasing mixing ratio of anionic and cationic surfactants. In particular, the pore structure of hexagonal lattice was found to strongly depend on the mixing ratio. Moreover, we can also control the size of mesopores by changing the chain length of hydrophobic parts of cationic surfactants. The method reported in this paper believed to be applicable to synthesize silica particles with highly regulated mesopores. Acknowledgments

Fig. 6. Pore-size distribution of MPS-B calculated from desorption branch.

This work was funded by the CLUSTER of Ministry of Education, Culture, Sports, Science and Technology, Japan. We thank to Mr. Takeshi Shimomura and Mr. Tsuyoshi Ohashi for their measurements of TEM and N2 adsorption isotherm. References

Fig. 7. Interlayer distances of silica-mixed surfactant products as a function of molar ratio of SOS: (", !) MTAB-SOS, (Q, P) CTAB-SOS, (2, 1) STABSOS; black plots: hexagonal phase, white plots: lamellar phase.

of anionic and cationic mixed surfactant system. If we use the cationic surfactant having longer hydrophobic chain such as STAB, the structure with longer d100 space was formed and the point of phase transition from hexagonal phase to lamellar phase was observed at lower concentration of SOS. These results indicate that, if the volume of hydrophobic parts increases with increasing alkyl chain length, the curvature of molecular assemblies decreases. Hence, these results strongly support that we can precisely control the interlayer distance and structure of composite material by changing not only the molecular ratio of cationic and anionic surfactants but also the alkyl chain length of cationic surfactants. 4. Conclusion We have developed a MT method using a cationic and anionic mixed surfactant system to obtain precisely controlled

[1] S. Iijima, Nature 354 (1991) 56. [2] S. Iijima, T. Ichihashi, Nature 363 (1993) 603. [3] C.T. Kresge, M.E. Leonowicz, W.J. Toth, J.C. Vartuli, J.S. Beck, Nature 359 (1992) 710. [4] J.S. Beck, J.C. Vartuli, W.J. Toth, M.E. Leonowicz, C.T. Kresgea, K.D. Schmitt, C.T.W. Chu, D.H. Olson, E.W. Sheppard, S.B. McCullen, J.B. Higgings, J.L. Schlenker, J. Am. Chem. Soc. 114 (1992) 10834. [5] S. Inagaki, Y. Fukushima, K. Kuroda, J. Chem. Soc. Jpn. 63 (1990) 988. [6] A. Monnier, F. Shuth, Q. Huo, D. Kumar, D. Margolese, R.S. Maxwell, G.D. Stucky, M. Krishnamurty, P. Petroff, A. Firouzi, M. Janicke, B.F. Chmelka, Science 261 (1993) 1299. [7] Q. Huo, D.I. Margolese, U. Ciesla, P. Feng, T.E. Gier, P. Sieger, R. Leon, P.M. Petroff, F. Schuth, G.D. Stucky, Nature 368 (1994) 317. [8] A.B. Stephen, E. Prouzet, T.J. Pinnavaia, Science 269 (1995) 1242. [9] S.A. Davis, S.L. Burkett, N.H. Mendelson, S. Mann, Nature 385 (1997) 420. [10] S.A. Bagshaw, T.J. Pinnavaia, Angew. Chem. Int. Ed. 35 (1996) 1102. [11] V. Luca, D.J. MacLachlan, J.M. Hook, R. Withers, Chem. Mater. 7 (1995) 2220. [12] Q. Huo, D.I. Margolese, G.D. Stucky, Chem. Mater. 8 (1996) 1147. [13] M.T. Yatcilla, K.L. Herrington, L.L. Brasher, E.W. Kaler, S. Chiruvolu, J.A. Zasadzinski, J. Phys. Chem. 100 (1996) 5874. [14] K. Tsuchiya, H. Nakanishi, H. Sakai, M. Abe, Langmuir 20 (2004) 2117. [15] H. Nakanishi, K. Tsuchiya, T. Ohkubo, H. Sakai, M. Abe, J. Oleo Sci. 54 (2005) 443. [16] D. Khushalani, A. Kuperman, N. Coombs, G.A. Ozin, Chem. Mater. 8 (1996) 2188. [17] L. Qi, J. Ma, H. Cheng, Z. Zhao, Chem. Mater. 10 (1998) 1623. [18] R. Ryoo, S.H. Joo, J.M. Kim, J. Phys. Chem. B 103 (1999) 7435. [19] L.R. Dai, T.W. Wang, L.T. Bu, G. Chen, Colloids Surf. A 181 (2001) 151. [20] F. Chen, F. Song, Q. Li, Micoporous Mesoporous Mater. 29 (1999) 305. [21] S. Pevzner, O. Regev, A. Lind, M. Lindén, J. Am. Chem. Soc. 125 (2003) 652. [22] A. Lind, B. Spliethoff, M. Lindén, Chem. Mater. 15 (2003) 813. [23] F. Chen, L. Huang, Q. Li, Chem. Mater. 9 (1997) 2685. [24] P.I. Ravikovitch, D. Wei, W.T. Chueh, G.L. Haller, A.V. Neimark, J. Phys. Chem. B 101 (1997) 3671. [25] P.I. Ravikovitch, A.V. Neimark, Colloids Surf. A 187–188 (2001) 11. [26] M. Thommes, B. Smarsly, M. Groenewolt, P.I. Ravikovitch, A.V. Neimark, Langmuir 22 (2006) 756. [27] D. Khushalani, A. Kuperman, G.A. Ozin, K. Tanaka, J. Garces, M.M. Olken, N. Coombs, Adv. Mater. 7 (1995) 842.