Preparation of core–shell mesoporous silica nanoparticles with bimodal pore structures by regrowth method

Preparation of core–shell mesoporous silica nanoparticles with bimodal pore structures by regrowth method

Accepted Manuscript Preparation of core–shell mesoporous silica nanoparticles with bimodal pore structures by regrowth method Hirotaka Ishii, Takaaki ...

1MB Sizes 5 Downloads 56 Views

Accepted Manuscript Preparation of core–shell mesoporous silica nanoparticles with bimodal pore structures by regrowth method Hirotaka Ishii, Takaaki Ikuno, Atsushi Shimojima, Tatsuya Okubo PII: DOI: Reference:

S0021-9797(15)00103-4 http://dx.doi.org/10.1016/j.jcis.2015.01.057 YJCIS 20199

To appear in:

Journal of Colloid and Interface Science

Received Date: Accepted Date:

20 November 2014 21 January 2015

Please cite this article as: H. Ishii, T. Ikuno, A. Shimojima, T. Okubo, Preparation of core–shell mesoporous silica nanoparticles with bimodal pore structures by regrowth method, Journal of Colloid and Interface Science (2015), doi: http://dx.doi.org/10.1016/j.jcis.2015.01.057

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Preparation of core–shell mesoporous silica nanoparticles with bimodal pore structures by regrowth method Hirotaka Ishii, Takaaki Ikuno, Atsushi Shimojima†, and Tatsuya Okubo*

Department of Chemical System Engineering, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8656, Japan, [email protected] †Present address: Department of Applied Chemistry, Waseda University, 3-4-1 Ohkubo, Shinjuku-ku, Tokyo 169-8555, Japan

Abstract Core–shell structured mesoporous silica nanoparticles (MSNs) with different pore characteristics in the cores and shells have been prepared by the regrowth method.

Adding a

silica source to a dispersion of presynthesized silica–surfactant composite nanoparticles with two-dimensional hexagonal mesostructures results in regrowth in preference to generation of new particles. Core–shell MSNs with bimodal porosities are easily obtained by adding a pore-expanding agent, 1,3,5-trimethylbenzene, in either the core or shell formation step. Detailed characterization of the core–shell MSNs reveals that the shells consist of disordered arrangements of relatively large or small pores and that the pore sizes in the cores change when the shells formed. Core–shell MSNs will be useful for controlling the release rates of the encapsulated guest molecules and for protecting internal pores from being plugged by other species.

Keywords: Mesoporous silica nanoparticles, core–shell structure, pore size control

1. Introduction Mesoporous silica nanoparticles (MSNs) have received considerable attention because of their wide range of potential applications such as catalysis, adsorption, drug delivery systems, 1

bioimaging, and anti-reflective coatings [1–6]. MSNs with tunable sizes and internal pore structures have been prepared by the reactions at low surfactant concentrations [7], by adding diols [8], by using different types of base catalysts [9–11], by using dual surfactants [12,13], and by the reactions in biphasic emulsion systems [10,11].

Organically modified MSNs

have also been prepared using organoalkoxysilanes as precursors [13–16]. core–shell structures are very important in advanced applications [17].

MSNs with

Core–shell MSNs

with metal and metal oxide cores have been prepared for use as nanocatalysts [18,19] and nanocarriers [3]. Selective functionalization of the inner and outer regions of the MSNs with different organosilanes has been achieved by the co-condensation method [20]. Core–shell nanoparticles can also be used for producing hollow MSNs by selective etching of the cores [21–23]. Core–shell MSNs with bimodal pore size distributions, i.e., with different pore sizes in cores and shells, are crucial in several applications. For example, relatively large pores increase the porosity of MSNs used as fillers in anti-reflective coatings [6], but they allow the polymer matrix to infiltrate the mesopores; this can be ameliorated using core–shell MSNs with smaller pores in the shell. Such core–shell MSNs are also useful as drug carriers for controlling the release rate while maintaining a high drug loading capacity. There have been a few studies on the successful preparation of such core–shell MSNs using the dual-templating method.

Areva et al. reported aerosol-assisted synthesis of core–shell

MSNs with a bimodal pore structure using a nonionic PEO-PPO-PEO-type triblock copolymer (F127) and cationic fluorocarbon surfactant as a co-template [24].

Niu et al.

synthesized core–shell structured MSNs with large pores in the cores and both large and small pores in the shells using an amphiphilic block copolymer (polystyrene-b-poly(acrylic acid)) and cetyltrimethylammonium bromide (CTAB) [25]. Most recently, Shen et al. reported the preparation of dendritic MSNs with radial mesopores, where the pore size of each generation can be adjusted using varied hydrophobic solvents in the oil–water biphasic systems [26]. Nonetheless, the variation of the pore characteristics of bimodal MSNs has been still limited. In addition, the previous methods have not been able to produce monodispersed MSNs of 2

diameters less than 100 nm that are desired for applications in drug delivery systems and nanocoatings. In this study, we demonstrate the preparation of relatively small (<100 nm), core–shell MSNs with different pore sizes in the cores and shells. The MSNs were formed through regrowth of silica–surfactant composite nanoparticles with a two-dimensional (2D) hexagonal structure (Scheme 1).

First, regrowth of silica–surfactant composite nanoparticles without

generating new particles was achieved by adding a silica source to the dispersion. Second, core–shell particles with smaller pores in the cores and larger pores in the shells were prepared by adding 1,3,5-trimethylbenzene (TMB) to act as a swelling agent in the shell formation step. Third, MSNs with larger pores in the cores and smaller pores in the shells were synthesized by adding TMB in the core formation step and by adding additional CTAB in the shell formation step. Although pore size expansion of mesoporous silica by adding pore-expanding agents such as TMB, alkylamines, and toluene has been reported [27–30], we propose a novel approach for the successful preparation of core–shell MSNs with bimodal pore structures.

3

Scheme 1. Three types of core–shell MSNs with (a) same pore sizes in the core and shell, (b) smaller pores in the core and larger pores in the shell, and (c) larger pores in the core and smaller pores in the shell.

Note that in reality, the shell regions in (b) and (c) have

disordered worm-like mesopores, but ordered pore arrangements are shown here for simplicity.

2. Experimental Section 2.1 Materials Ethylene glycol (EG) (Wako Pure Chemical), hexadecyltrimethylammonium bromide (Wako), 25% ammonia solution (Wako), 1,3,5-trimethylbenzene (Tokyo Chemical Industry (TCI)), tetraethyl orthosilicate (TEOS; Wako), 3-aminopropyltriethoxysilane (APTES; TCI), 2-propanol (Wako), ethanol (Wako), 5 N HCl (Wako), and hexamethyldisiloxane (HMDS; TCI) were used as received.

2.2 Preparation of MSNs with controlled sizes using the regrowth method A mixture of EG, NH3, CTAB, and water was stirred at 333 K for 30 min. A mixture of TEOS and APTES was then added slowly with vigorous stirring at 333 K. The mixture was stirred at 333 K for 4 h, and an opaque dispersion of silica–surfactant composite nanoparticles 4

(SSNs) was obtained. Regrowth of these “seed” nanoparticles was achieved by adding a TEOS–APTES mixture to the dispersion and stirring at 333 K for 1 h. This regrowth procedure was performed three times.

The amounts of TEOS and APTES added in each

regrowth step were the same as those used for the preparation of the seed particles. The molar compositions of the final mixtures are shown in Table 1. The as-synthesized SSNs were collected for characterization by centrifugation.

These samples are hereafter called SSN1-n,

where n is the number of regrowth steps performed (n = 0, 1, 2, or 3).

The surfactant was

removed using the Lentz method. An as-synthesized dispersion was slowly added to a biphasic mixture of HMDS (13 g), 5 N HCl (30 g), and 2-propanol (15 g) that had been previously stirred at 345 K for 30 min. Under these conditions, cleavage of HMDS and exchange of the surfactant cations with protons proceed, resulting in trimethylsilylation of the silanol groups on both the internal and external surfaces of SSNs [6,31]. Trimethylsilylated MSNs were collected by centrifuging the organic phase, and the MSNs were washed with 2-propanol and air dried at 333 K. The MSN samples are hereafter called MSN1-n.

Table 1. The final molar compositions for preparing SSNs Sample*

TEOS

APTES

CTAB

NH3

EG

H2 O

SSN1-0/MSN1-0

1.00

0.17

1.06

12.8

52

1075

SSN1-1/MSN1-1

2.00

0.34

0.53

12.8

52

1075

SSN1-2/MSN1-2

3.00

0.51

0.53

12.8

52

1075

SSN1-3/MSN1-3

4.00

0.68

0.53

12.8

52

1075

*The reagents used for trimethylsilylation are not shown here.

2.3 Preparation of core–shell MSNs with larger pores in the shells than in the cores Core–shell MSNs with larger pores in the shells than in the cores were synthesized by adding a pore expanding agent, TMB, in the regrowth step (Scheme 1(b)). Seed SSNs were prepared as described in section 2.2, and TMB was added to this dispersion.

The mixture

was stirred for 30 min, and regrowth was achieved by adding a TEOS–APTES mixture, 5

followed by stirring at 333 K for 1 h. The TMB/CTAB molar ratio was 4, and the amounts of TEOS and APTES added for the regrowth step were the same as those used for synthesizing SSN1-0. The molar compositions of the final mixtures are listed in Table 2. The samples before and after regrowth were trimethylsilylated by the Lentz method; the resulting samples are hereafter called MSN2-0 and MSN2-1, respectively.

2.4 Preparation of core–shell MSNs with larger pores in the cores than in the shells Seed SSNs with larger pores in the cores than in the shells were synthesized using TMB, and more CTAB was added during the regrowth process to decrease the TMB/CTAB ratio in the liquid phase of the SSN dispersion (Scheme 1(c)). A mixture of EG, NH3, and CTAB was stirred at 333 K for 30 min; then TMB was added under stirring. A mixture of TEOS and APTES was then added slowly to the solution, with vigorous stirring at 333 K, and the mixture was stirred for 4 h to produce seed SSNs.

Regrowth was achieved by first adding

CTAB to the seed dispersion and stirring the mixture for 30 min.

A TEOS–APTES mixture

was then added, and the mixture was stirred for 2 h. The amount of CTAB present was seven times higher than that used for synthesizing the seed SSNs. The molar compositions of the final mixtures are shown in Table 2. The resulting samples were trimethylsilylated by the Lentz method and are hereafter called MSN3-0 (before regrowth) and MSN3-1 (after regrowth). Table 2. The final molar compositions for preparing core-shell-type SSNs Sample*

TEOS

APTES

CTAB

NH3

EG

H2O

TMB

MSN2-0

1.00

0.17

0.53

12.8

52

1075

0

MSN2-1

2.00

0.34

0.53

12.8

52

1075

2.12

MSN3-0

1.00

0.17

0.53

12.8

52

1075

2.12

MSN3-1

2.00

0.34

3.71

12.8

52

1075

2.12

*The reagents used for trimethylsilylation are not shown here.

6

2.5 Characterization Powder X-ray diffraction (XRD) patterns were collected using M03X-HF (Bruker AXS) with Cu–Kα radiation. Fourier transform infrared (FT-IR) spectra were recorded on Nicolet Magna 560 FTIR (Thermo) by the KBr pellet technique.

Field-emission scanning electron

microscope (FE-SEM) images were obtained using S-900 (HITACHI) at 6 kV after coating with Pt for 15 sec using an ion sputter system E-1030 (Hitachi) with a magnetron electrode. Transmission electron microscope (TEM) images were collected using JEM-2000EX (JEOL) operated at 200 kV. Nitrogen adsorption–desorption measurements were performed at 77 K using Autosorb-1-MP (Quantachrome Instrument) after outgassing the samples at 425 K for 6 h. Pore size distributions were calculated by Barrett-Joyner-Halenda (BJH) method using adsorption branches of the isotherms. Non-localized density functional theory (NLDFT) was also applied to calculate the pore size distribution (Fig. S1 in the Supplementary Information). N2-silica adsorption branch kernel at 77 K based on a cylindrical pore model was used. Thermogravimetry (TG) was performed on PU 4K (Rigaku) using a 10% O2:90% He atmosphere at a heating rate of 10 K min−1. Dried samples were characterized using XRD, TG, CHN elemental analysis, nitrogen adsorption–desorption isotherms, and FT-IR spectroscopy.

Dispersed particles were coated on Si wafers and Cu TEM grids for SEM and

TEM observations, respectively.

3

Results and Discussion

3.1 Verification of regrowth of silica–surfactant composite nanoparticles Regrowth of spherical particles of mesostructured silica–surfactant composites has been reported for particles with radially oriented channels [26, 32] or disordered worm-like structures [33].

Here, we investigated regrowth of nanoparticles with a well-ordered 2D

hexagonal mesostructure with straight channels running through the particles. Fig. 1 shows the FE-SEM images of the silica–surfactant composite nanoparticles before regrowth (SSN1-0), and those after one, two, and three regrowth procedures had been performed (SSN1-1, SSN1-2, and SSN1-3, respectively). 7

The average size of the seed

particles (SSN1-0) was 59 nm in diameter. The size of the particles increased with the number of regrowth procedures, and particle diameters reached ca. 94 nm after three regrowth procedures. The increase in particle diameter corresponds to a fourfold increase in particle volume.

This indicates that the silicate species added to the dispersion were mainly

consumed for the growth of the seed SSNs.

However, the small number of tiny particles and

irregularly shaped large particles that can be observed in Figs. 1(a–c) suggests that spontaneous nucleation and aggregation also occurred. We expect that the regrowth process could be controlled more precisely by optimizing the reaction conditions.

For example,

Yamamoto and co-workers recently reported the formation of highly monodispersed MSNs via seed growth [33].

The key to this method is using tetrapropoxysilane (TPOS) rather than

TEOS so that concentration of the silicate species can be maintained below the critical nucleation concentration (because TPOS has a lower hydrolysis rate compared with TEOS).

Fig. 1. FE-SEM images of (a) SSN1-0, (b) SSN1-1, (c) SSN1-2, and (d) SSN1-3.

The

samples were treated with aqueous HCl to remove surfactants before the images were acquired.

The compositions of SSNs recovered by centrifugation were evaluated by TG and CHN analyses (Table S1 in Supplementary Information). From the weight and SiO2 content, the yield of SSN1-0 based on SiO2 was found to be 44%. The yield increased to 72% after the 8

first regrowth procedure (i.e., for SSN1-1). The difference between the yields was caused by the amount of dissolved silicate species in the liquid phase being almost the same before and after regrowth.

Although over half the silica source initially added remained dissolved in the

liquid phase under our experimental conditions, most of the silica source subsequently added was consumed for the regrowth. Approximately 83% of CTAB remained in the liquid phase after the seed particles (SSN1-0) had formed.

This value decreased to 63% after the first

regrowth procedure, meaning that the regrown part (shell) consisted of silica and CTA cations, forming a mesostructured silica–surfactant composite. The FT-IR analyses confirmed that the surfactant template could be removed by the Lentz method although the particles were coated with shells (data not shown).

Fig. 2 shows the

XRD patterns for samples MSN1-0 and MSN1-1. The peaks assigned to a 2D hexagonal (p6mm) structure with a d10 spacing of 4.39 nm were observed for MSN1-0 (Fig. 2(a)).

A

similar pattern with slightly broadened peaks was observed for MSN1-1, suggesting that the shells also had 2D hexagonal structures. The TEM image of MSN1-0 shows that spherical particles with hexagonal dot or striped patterns were observed (Figs. 3(a, b)). These patterns could also be seen near the surfaces of the MSN1-1 particles (Figs. 3(c, d)). The average pore size obtained from nitrogen adsorption measurements was almost the same (2.5 nm) before and after regrowth (Table 4). These results indicate that epitaxial-like growth of 2D hexagonal mesostructures occurred during regrowth.

9

Fig. 2. XRD patterns of (a) MSN1-0 and (b) MSN1-1.

10

Fig. 3. TEM images of (a), (b) MSN1-0 and (c), (d) MSN1-1.

11

Scale bars: 50 nm.

Table 3. Pore characteristics of MSNs. Sample

BET surface area

Pore volume

BJH Pore

[m2 g−1]

[cm3 g−1]

diameter [nm]

MSN1-0

880

1.1

2.4

MSN1-1

1000

1.4

2.5

MSN2-0

1100

1.4

2.5

MSN2-1

800

1.5

3.1, 4.5

MSN3-0

550

1.1

4.7

MSN3-1

670

1.2

2.8, 3.7

3.2 Preparation of core–shell MSNs with larger pores in the shells than in the cores Fig. 4 shows the FE-SEM images of the core MSNs (MSN2-0) and the product after regrowth in the presence of TMB (MSN2-1). Note that MSN2-0 was prepared under the conditions identical to those for MSN1-0, but a newly prepared sample was used for an accurate comparison with MSN2-1. The increase of the average particle sizes suggests that regrowth occurred. The TEM image of MSN2-1 (Fig. 5) shows that the particles have relatively dark inner parts and relatively bright outer parts, which suggests that the particles have core–shell structures. The shells appear to have wormhole-like structures, which are different from the well-ordered hexagonal structures formed in the absence of TMB (Fig. 3). The XRD patterns of the samples before and after regrowth are shown in Fig. 6.

The

regrown sample exhibits an additional, relatively broad peak (d = 7.1 nm) characteristic of the disordered, wormhole-like structures at the lower 2θregion of the d10 peak for the cores (Fig. 6b). These results strongly suggest that mesostructured shells with larger periodicities were formed on the core particles.

12

Fig. 4. FE-SEM images of (a) MSN2-0 and (b) MSN2-1.

13

Fig. 5. TEM image of MSN2-1.

14

Fig. 6. XRD patterns of (a) MSN2-0 and (b) MSN2-1.

15

Nitrogen adsorption–desorption isotherms and BJH pore size distribution curves for MSN2-0 and MSN2-1 are shown in Fig. 7.

The BET surface areas, pore volumes, and pore

diameters are listed in Table 3. After regrowth, an additional capillary condensation step appeared (P/P0 = ~0.5) in the isotherm, and two peaks were observed at 3.1 and 4.5 nm in the BJH pore size distribution (Fig. 7(b)). The hysteresis at P/P0 = 0.8–1.0 observed for both samples was attributed to interparticle voids. We concluded that core–shell MSNs with smaller pores in the cores and larger pores in the shells were successfully synthesized by the regrowth method in the presence of TMB.

The increase in the pore size in the cores (from

2.5 to 3.1 nm) is consistent with the slight increase in the d10 spacing (Fig. 6). These results suggest that TMB acted as a pore expander for the as-synthesized SSNs, which was accompanied by rearrangement of the silicate frameworks.

In fact, the pore size increased

when only TMB was added to the SSN dispersion, and the mixture was aged at 333 K for 4 h (data not shown).

Such pore size expansions during post-treatment of mesostructured

silica–surfactant composites in the presence of swelling agents were reported [34, 35]. Larger increases in pore sizes than that observed in our study were achieved by performing post-treatments at higher temperatures and using longer aging times (e.g., under hydrothermal conditions for several days).

Therefore, in our system, it is important to optimize the

reaction conditions for the shell formation step to minimize the increase in pore size in the cores.

16

Fig. 7. Nitrogen adsorption–desorption isotherms (left) and BJH pore size distribution curves (right) for (a) MSN2-0 and (b) MSN2-1.

3.3 Preparation of core–shell MSNs with larger pores in the cores than in the shells The XRD pattern for the seed particles prepared in the presence of TMB (MSN3-0) shows several peaks that are attributed to a 2D hexagonal structure with a d10 spacing of 7.2 nm (Fig. 8(a)). The product of the regrowth procedure in the presence of the additional CTAB, MSN3-1, has a broadened peak at d = 6.4 nm (Fig. 8(b)). A small shoulder also appears at a higher 2θ, which suggests the presence of a different phase. Fig. 9 shows the TEM images of MSN3-0 and MSN3-1. Cylindrical pores open to the outer surfaces are clearly seen in the particles before regrowth, whereas disordered pores are found in the outer parts after regrowth; although hexagonally arranged pores are still observed at the inner regions (Figs. 9(c, d)). The nitrogen adsorption–desorption isotherms and BJH pore size distribution curves for MSN3-0 and MSN3-1 are shown in Fig. 10. The BET surface areas, pore volumes, and pore diameters are listed in Table 3. MSN3-0 has a narrow pore size distribution centered at 4.7 nm, which is much larger than the pore size of the particles produced without adding TMB (ca. 2.5 nm for MSN1-0 and MSN2-0). However, MSN3-1 shows two peaks at 3.7 and 2.8 nm. Hence, it is reasonable to conclude that shells with smaller pores (2.8 nm) were formed, and that the average pore size in the cores decreased to 3.7 nm. The decrease in pore size in the cores during the regrowth procedure is probably due to the shrinkage of the silicate 17

framework and/or to the deposition of additional silicate species on the pore walls.

Such a

bimodal pore size distribution was not achieved when CTAB was not added during regrowth (data not shown). Therefore, the addition of CTAB plays an important role in decreasing the micelle size, allowing core–shell MSNs with larger pores in the cores and smaller pores in the shells to be formed.

Fig. 8. XRD patterns of (a) MSN3-0 and (b) MSN3-1.

18

Fig. 9. TEM images of (a), (b) MSN3-0 and (c), (d) MSN3-1.

19

Scale bars: 50 nm.

Fig. 10. Nitrogen adsorption–desorption isotherms (left) and BJH pore size distribution curves (right) for (a) MSN3-0 and (b) MSN3-1.

4.

Conclusions We developed a facile method, based on regrowth of silica–surfactant composite

nanoparticles, for producing core–shell MSNs with bimodal pore structures.

The addition of

a silica source to the dispersion of core nanoparticle allowed regrowth to occur in preference to the generation of new particles.

Core–shell nanoparticles with smaller mesopores in the

cores and larger mesostructures in the shells were obtained by adding TMB as a pore-expander in the regrowth procedure. Core–shell nanoparticles with larger pores in the cores and smaller pores in the shells were obtained by adding TMB in the core formation step and by adding CTAB in the shell formation step. The core–shell MSNs could be useful for fabricating low-n and low-k materials and for efficiently controlling the adsorption and release rates of the guest molecules.

Acknowledgements The authors thank Dr. Yasuto Hoshikawa (Tohoku University) and Prof. Ayae Sugawara-Narutaki (Nagoya University) for fruitful discussion.

This work was partially

supported by a Grant-in-Aid for Scientific Research (B) and a Grant-in Aid for Scientific Research (C) from the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan. A part of this work was conducted at the Center for Nano Lithography & 20

Analysis, The University of Tokyo, supported by MEXT.

References [1]S.-H. Wu, C.-Y. Moua, H.-P. Lin, Chem. Soc. Rev. 42 (2013) 3862. [2]I. I. Slowing, B. G. Trewyn, S. Giri, V. S.-Y. Lin, Adv. Funct. Mater. 17 (2007) 1225. [3]F. Tang, L. Li, D. Chen, Adv. Mater. 24 (2012) 1504. [4]P. Yang, S. Gai, J. Lin, Chem. Soc. Rev. 41 (2012) 3679. [5]M. Yamaguchi, H. Nakayama, K. Yamada, H. Imai, Opt. Lett. 34 (2009) 2060. [6]Y. Hoshikawa, H. Yabe, A. Nomura, T. Yamaki, A. Shimojima, T. Okubo, Chem. Mater. 22 (2010) 12. [7]Q. Cai, Z. S. Luo, W. Q. Pang, Y. W. Fan, X. H. Chen, F. Z. Cui, Chem. Mater. 13 (2001) 258. [8]J. Gu, W. Fan, A. Shimojima, T. Okubo, Small. 3 (2007) 1740. [9] K. Moller, J. Kobler, T. Bein, Adv. Funct. Mater. 17 (2007) 605. [10] T. Yokoi, T. Karouji, S. Ohta, J. N. Kondo, T. Tatsumi, Chem. Mater. 22 (2010) 3900. [11] J. Wang, A. Sugawara-Narutaki, A. Shimojima, T. Okubo, J. Wang, A. Sugawara-Narutaki, A. Shimojima, T. Okubo, J. Colloid Interface Sci. 385 (2012) 41. [12] K. Suzuki, K. Ikari, H. Imai, J. Am. Chem. Soc. 126 (2004) 462. [13] J. Gu, K. Huang, X. Zhu, Y. Li, J. Wei, W. Zhao, C. Liu, J. Shi, J. Colloid Interface Sci. 407 (2013) 236. [14] Y.-S. Lin, N. Abadeer, K. R. Hurley, C. L. Haynes, J. Am. Chem. Soc. 133 (2011) 20444. [15] J. Kobler, K. Mo ller, T. Bein, ACS Nano 2 (2008) 791. [16] C. Urata, H. Yamada, R. Wakabayashi, Y. Aoyama, S. Hirosawa, S. Arai, S. Takeoka, Y. Yamauchi, K. Kuroda, J. Am. Chem. Soc. 133 (2011) 8102. [17] B. J. Jankiewicz, D. Jamiola, J. Choma, M. Jaroniec, Adv. Colloid Interface Sci. 170 (2012) 28. [18] S. H. Joo, J. Y. Park, C. K. Tsung, Y. Yamada, P. D. Yang, G. A. Somorjai, Nat. Mater. 8 (2009) 126. 21

[19] Q. Zhang, I. Lee, J. B. Joo, F. Zaera, Y. Yin, Acc. Chem. Res. 46 (2013) 1816. [20] V. Cauda, A. Schlossbauer, J. Kecht, A. Zurner, T. Bein, J. Am. Chem. Soc. 131 (2009) 11361. [21] H. Blas, M. Save, P. Pasetto, C. Boissiere, C. Sanchez, B. Charleux, Langmuir 24 (2008) 13132. [22] Y. Chem, H. Chen, L. Guo, Q. He, F. Chen, J. Zhou, J. Feng, J. Shi, ACS Nano 4 (2010) 529. [23] T. Ikuno, A. Nomura, K. Iyoki, A. Sugawara-Narutaki, T. Okubo, A. Shimojima, Chem. Lett. 42 (2013) 316. [24] S. Areva, C. Boissière, D. Grosso, T. Asakawa, C. Sanchez, M. Lindén, Chem. Commun. (2004) 1630. [25] D. C. Niu, Z. Ma, Y. S. Li, J. L. Shi, J. Am. Chem. Soc. 132 (2010) 15144. [26] D. Shen, J. Yang, X. Li, L. Zhou, R. Zhang, W. Li, L. Chen, R. Wang, F. Zhang, D. Zhao, Nano Lett. 14 (2014) 923. [27] K.-C. Kao, C.-Y. Mou, Microporous Mesoporous Mater. 169 (2013) 7. [28] G. Sponchia, R. Marin, I. Freris, M. Marchiori, E. Moretti, L. Storaro, P. Canton, A. Lausi, A. Benedetti, P. Riello, J. Nanopart. Res. 16 (2014) 2245. [29] A. B. Fuertes, P. Valle-Vigón, M. Sevilla, J. Colloid Interface Sci. 349 (2010) 173. [30] J. Peng, J. Liu, J. Liu, Y. Yang, C. Li, Q. Yang, J. Mater. Chem. A 2 (2014) 8118. [31] C.W. Lentz, Inorg. Chem., 3 (1964) 574. [32] T. Nakamura, M. Mizutani, H. Nozaki, N. Suzuki, K. Yano, J. Phys. Chem. C. 111 (2007) 1093. [33] E. Yamamoto, M. Kitahara, T. Tsumura, K. Kuroda, Chem. Mater. 26 (2014) 2927. [34] A. Sayari, M. Kruk, M. Jaroniec, I. L. Moudrakovski, Adv. Mater. 10 (1998) 1376. [35] M. Mizutani, Y. Yamada, K. Yano, Chem. Commun. (2007) 1172.

22

Graphical Abstract

Core–shell structured mesoporous silica nanoparticles with different pore characteristics in the cores and shells have been prepared by the regrowth method in which 1,3,5-trimethylbenzene was added as a pore-expander in either the core or shell formation step.

23

Highlights

• A facile method for producing core–shell-type mesoporous silica nanoparticles with bimodal pore structures is developed. • Core–shell nanoparticles with smaller mesopores in the cores and larger mesostructures in the shells are obtained by adding pore-expander in the regrowth procedure. • Core–shell nanoparticles with larger pores in the cores and smaller pores in the shells were obtained by adding pore-expander in the core formation step and by adding CTAB in the shell formation step.

24