DOI: 10.1002/chem.201501599
Full Paper
& Silica Materials
A New Synthetic Route to Microporous Silica with Well-Defined Pores by Replication of a Metal–Organic Framework Atsushi Kondo,*[a] Anthony Shoji Hall,[b] Thomas E. Mallouk,[b] and Kazuyuki Maeda[a] Abstract: Microporous amorphous hydrophobic silica materials with well-defined pores were synthesized by replication of the metal–organic framework (MOF) [Cu3(1,3,5-benzenetricarboxylate)2] (HKUST-1). The silica replicas were obtained by using tetramethoxysilane or tetraethoxysilane as silica precursors and have a micro–meso binary pore system. The BET surface area, the micropore volume, and the mesopore volume of the silica replica, obtained by means of hydrothermal treatment at 423 K with tetraethoxysilane, are
Introduction Inorganic oxides are important materials not only because of their mechanical and chemical stability, but also because of their versatile functionality as adsorbents, catalysts and supports, and as ferroic, magnetic, and dielectric materials. Silica materials are of particular interest for applications in molecular adsorption, separation, and drug delivery, and these materials provide a useful platform on which to covalently anchor various functionalities.[1–5] The broad scope of silica chemistry is largely based on the facile formation of its network structure, which can take various micro- and mesoporous forms, and the robust chemistry that exists for its surface modification. Mesoporous silica represents a successful example of nanoscale structural control that is achieved by means of colloid chemistry and sol–gel chemistry.[6–9] In the assembly of mesoporous silica materials, surfactant species with different chain lengths result in different pore sizes and network connectivities. The resulting materials have adsorption and catalytic properties that strongly correlate to the pore dimensions and surface chemistry. However, soft template structures are often flexible and can be dependent on temperature, solvent, charge of the ions, and other parameters, which makes the prediction of the [a] Dr. A. Kondo, Prof. K. Maeda Department of Applied Chemistry, Graduate School of Engineering, Tokyo University of Agriculture and Technology 2-24-16 Naka-cho, Koganei, Tokyo 184-8588 (Japan) Fax: (+ 81) 42-388-7040 E-mail: [email protected] [b] Dr. A. S. Hall, Prof. T. E. Mallouk Department of Chemistry The Pennsylvania State University 104 Chemistry Building, University Park, PA 16802 (USA) Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/chem.201501599. Chem. Eur. J. 2015, 21, 12148 – 12152
620 m2g¢1, 0.18 mL g¢1, and 0.55 mL g¢1, respectively. Interestingly, the silica has micropores with a pore size of 0.55 nm that corresponds to the pore-wall thickness of the template MOF. The silica replica is hydrophobic, as confirmed by adsorption analyses, although the replica has a certain amount of silanol groups. This hydrophobicity is due to the unique condensation environment of the silica precursors in the template MOF.
resulting negative replica more difficult; therefore, other synthetic methodologies have been investigated to broaden the scope of silica chemistry. One promising route to oxide materials with well-defined pore networks is through the replication of ordered materials that act as hard templates.[10–14] For example, spherical particles are used in the colloidal-crystal templating process, which enables the synthesis of many different types of replica materials with three-dimensional porosity on length scales of tens of nanometers to microns.[15–20] The pore connectivity and dimensions of replica materials derive from the structural characteristics of the template. For example, zeolites can be utilized to fabricate microporous carbon materials of extremely high surface area, and the structure and porosity depends on the dimensionality of the original zeolite pore network.[21–25] Recently, metal–organic frameworks (MOFs) or porous coordination polymers, which are composed of metal ions and organic linkers, have emerged as promising materials for gas storage/separation, sensing, and catalysis.[26–34] Because MOFs have a variety of structures with different pore sizes, shapes, and network connectivities, these materials are good candidates as hard templates for fabricating unique materials with micro- and mesoporous structures.[35–41] Very recently, we demonstrated this idea by making microporous titania replicas of the coppercontaining cubic MOF [Cu3(1,3,5-benzenetricarboxylate)2] (HKUST-1).[42, 43] This MOF provides an ordered nanospace that can withstand the conditions of a hydrothermal reaction, but is soluble enough to be removed under very mild conditions. Herein, we explore the synthesis of silica replicas of this MOF. Although the pore size of the MOF is close to the size of the silica precursor, the precursor molecules infiltrate into the micropores and form a silica network in a hydrothermal reaction. The final silica product is not crystallographically ordered, but has well-defined micropores with a pore size that corre-
12148
Ó 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Full Paper sponds to the pore-wall thickness of the template material. Interestingly, adsorption experiments show that the resulting microporous silica materials are hydrophobic. The success of the MOF replication method with silica suggests that it may be possible to generalize the method to other oxides and perhaps to other MOF templates, thus leading to materials of versatile compositions and structures.
Results and Discussion XRD patterns of the HKUST-1 template and HKUST-1/silica composites obtained with tetraethoxysilane (TEOS) as the silica source after hydrothermal treatment are shown in Figure 1 a. The composites retain the structure of the HKUST-1 template up to 423 K. At higher hydrothermal-treatment temperatures (453 and 473 K), prominent diffraction lines at 36.4 and 42.38 2q appear in the XRD pattern, thus signaling the formation of Cu2O. Therefore, HKUST-1 can act as a hard template up to 423 K without decomposition to form silica composites. It is interesting that HKUST-1 retains its structure at higher temperatures (up to 473 K) when tetramethoxysilane (TMOS) was used instead as the silica precursor (see Figure 1 S in the Supporting Information), possibly because of the higher ceramic fraction of silica in TMOS relative to TEOS, which means the smaller contraction of TMOS molecules than TEOS through the condensation reaction to form the silica replica. Figure 1 b shows XRD patterns of the silica replicas obtained after the removal of the HKUST-1 template by HCl etching. The XRD patterns in Figure 1 b indicate that the silica replicas are amorphous and that the microcrystalline Cu2O is completely removed by the HCl etching. A broad amorphous halo peak was observed in the range of 20–308 for the sample treated at 423 K. This peak is weaker for the samples treated at higher temperatures, which may imply a decrease in short-range ordering of the
Figure 1. a) XRD patterns of the HKUST-1 template and the HKUST-1/silica composites obtained with TEOS as the silica source after hydrothermal treatment in the range of 353–473 K. The asterisk symbol (*) indicates diffraction peaks of Cu2O. b) XRD patterns of silica materials obtained after HKUST-1 removal. Chem. Eur. J. 2015, 21, 12148 – 12152
www.chemeurj.org
SiO4 tetrahedra. The temperature of the hydrothermal treatment was critical for the formation of silica networks in the pores of HKUST-1 because no solid was recovered after HCl etching when the composites made from TEOS were treated at 353 or 393 K. In contrast, the use of TMOS as a precursor gave solid silica products over a wide range of hydrothermal temperatures (i.e., 393–473 K). The silica replica obtained by hydrothermal treatment at 423 K by using TEOS was selected as a representative and was characterized. The energy-dispersive X-ray (EDX) spectrum of the sample predominantly shows the presence of silicon atoms and the absence of metal species, including copper, which is the metal ion in the HKUST-1 template (see Figure 2 S in the Supporting Information). The curve obtained by thermogravimetry (TG) of the etched material shows a gradual weight loss of 5 wt % in the temperature range 373–1273 K, thus indicating almost complete removal of organic moieties from the HKUST-1 template (see Figure 3 S in the Supporting Information). A solid-state 29Si cross-polarization magic-angle spinning (CP-MAS) NMR spectrum of the silica replica exhibited a peak at d = ¢100 ppm, which is assigned to the Q3 site with peaks at d = ¢91 and ¢111 ppm (Q2 and Q4 sites, respectively; Qn = Si(OT)n(OH)4¢n ; see Figure 4 S in the Supporting Information). Although it is not possible to conduct a quantitative analysis of the results obtained by NMR spectroscopic analysis because of the cross-polarization technique used, it is clear that the silica replica contains a considerable amount of Q3 units in addition to fully connected Q4 units and partially polymerized Q2 units. A mixture of silicon connectivity that ranges from Q2 to Q4 is also typically observed in amorphous porous silica materials, such as SBA-15,[44] thus indicating that the local silica network structure in the silica replica of HKUST-1 is similar to that of the amorphous silica materials. Figure 2 shows FTIR spectra of the etched material together with the HKUST1 template and the composite after hydrothermal treatment. The IR spectrum of the composite indicates the presence of 1,3,5-benzenetricarboxylate (BTC) ligands because of the strong absorption bands at n˜ = 1374 and 1647 cm¢1 , which are assignable to the symmetric and asymmetric stretching modes, respectively, of a carboxylate group.[45] The absence of absorption bands in the region n˜ = 1720–1700 cm¢1 suggests the absence of free carboxylic acid. Therefore, it is reasonable
Figure 2. FTIR spectra of the HKUST-1 template (upper), the composite after hydrothermal treatment (middle), and the silica replica obtained by HCl etching (bottom). The absorption bands of the carboxylate groups of the BTC ligand (+ +) and the asymmetric Si¢O bending mode of the SiO4 tetrahedra (*) are indicated.
12149
Ó 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Full Paper to suggest that coordination bonding between the carboxylate groups of the BTC ligands and copper(II) ions was retained throughout the hydrothermal treatment. The IR spectrum of the etched sample has no absorption bands assigned to BTC ligands, and an obvious broad peak in the range n˜ = 1000– 1200 cm¢1 is assigned to the asymmetric Si¢O bending mode of the SiO4 tertrahedra. In the spectrum of the etched material, a small broad absorption band in the region n˜ = 3600– 3100 cm¢1 is assignable to the stretching modes of the hydroxy groups, thus suggesting a certain amount of silanol groups (see Figure 5 S in the Supporting Information).[46, 47] In addition, the etched material shows no apparent absorption that corresponds to the C¢H vibration of alkyl chains in the range n˜ = 3000–2800 cm¢1, thus indicating the absence of alkoxy groups from the silica precursor molecules, which is consistent with the solid-state 13C MAS NMR results.[48] Many particles of the HKUST-1/silica composites retain the bipyramidal morphology of the parent MOF crystals, and no apparent silica particles were observed outside of HKUST-1 , thus suggesting that only negligible amounts of solid silica formed through the condensation reactions of the precursor molecules (see Figure 6 S in the Supporting Information). Figure 3 shows SEM images of the silica replica obtained by etching of the composite material. Although the etched silica materials do not perfectly retain the macroscale morphology of the MOF crystals, the particles have flat faces with sizes in the micrometer range, thus suggesting replication of the MOF template. A SEM image at high magnification (Figure 3 b) shows that the silica particles have fine pores in the range of mesopores (n = pore size; n = 2–50 nm) and macropores (n > 50 nm) formed from continuously connected silica networks.
Figure 3. a) A SEM image of the silica obtained by etching of the composite and b) an enlarged image of the boxed part of (a). Chem. Eur. J. 2015, 21, 12148 – 12152
www.chemeurj.org
Figure 4. a) Nitrogen adsorption–desorption isotherms of the silica replica at 77 K. Solid and open symbols represent adsorption and desorption branches, respectively. b) Pore-size distribution of the silica derived from the N2 adsorption isotherm at 77 K by using the HK method. In y axis, w and w0 means the pore volume and the micropore volume, respectively.
The adsorption properties of the silica replicas were examined to determine the pore-size distribution and pore volume. Figure 4 shows N2 adsorption–desorption isotherms of the silica material at 77 K. As expected from the SEM images, there is a hysteresis loop, which indicates the presence of mesopores. Because of the gradual increase over a wide range of relatively high pressures, the mesopore size is not uniform. This finding was confirmed by means of a Barrett–Joyner–Halenda (BJH) pore-size analysis[49] of the N2 adsorption isotherm (see Figure 7 S in the Supporting Information). A steep uptake at low relative pressure indicates the presence of micropores. The micropore-size distribution of the silica derived from the isotherm by using the Horvath–Kawazoe (HK) method[50] is shown in Figure 4 b. A sharp peak at 0.55 nm, which corresponds to the pore-wall thickness in the HKUST-1 template, is observed in the distribution function, although the function has a relatively broad distribution of micropore diameters. The micropore volume and isosteric heat of adsorption qst,f= 1/e evaluated by using the Dubinin–Radushkevich analysis[51] are 0.22 mL g¢1 and 11.0 kJ mol¢1, respectively. The BET surface area[52] of the silica was 620 m2 g¢1 and total pore volume was 0.77 mL g¢1. A mesopore volume of 0.55 mL g¢1 was obtained by subtracting the micropore volume from the total pore volume. Mesopores should be formed by incomplete filling of the HKUST-1 template with the silica component. Adsorption properties of other small molecules were also investigated. Figure 5 shows vapor-adsorption isotherms of water, methanol, ethanol, and tetrachloromethane on the silica
12150
Ó 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Full Paper surface of the silica replica to be hydrophobic. Copper(II) ions, which are arranged periodically in the template MOF, may have a catalytic influence to induce the condensation of the surface functional groups to form the hydrophobic pore surface. Thus, it should be noted that MOF templates have the potential to define not only the structural features, such as pore size and pore-network connectivity, but also the pore-surface properties.
Conclusion Figure 5. Adsorption isotherms of water (*), methanol (~), ethanol (&), and tetrachloromethane (^) on the obtained silica measured at 303 K. The solid and open symbols represent the adsorption and desorption branches, respectively.
replicas measured at 303 K. Relative to the methanol and ethanol isotherms, the water-adsorption isotherm shows a gentle increase in uptake over a wide relative pressure range with a smaller uptake amount. This behavior suggests that the material is relatively hydrophobic, which is consistent with the lack of silanol groups in the FTIR spectra. In contrast, the methanol- and ethanol-adsorption isotherms have steep uptakes in the low-pressure region and are very similar to each other. The amounts of adsorbed water, methanol, ethanol, and tetrachloromethane near P/P0 = 1 are 10.1, 16.3, 10.5, and 7.4 mmol g¢1, respectively. The ratio of the molecular volumes is 1:2.6:4.1:7.5 (water/methanol/ethanol/tetrachloromethane); therefore, the ratio of occupied volumes by the molecular species is 1.0:4.3:4.2:5.5 (water/methanol/ethanol/tetrachloromethane). This result clearly indicates that water molecules only partially occupy the pores of the silica, whereas methanol, ethanol, and tetrachloromethane occupy four to five times as much of the pore volume. The pore volumes evaluated for different molecular species are 0.18, 0.67, 0.61, and 0.71 mL g¢1, which are 23, 87, 79, and 92 % of the total pore volume estimated by N2 adsorption, for water, methanol, ethanol, and tetrachloromethane, respectively. The high filling percentage of tetrachloromethane directly indicates that the pore space in the silica replica is hydrophobic. The replicas made from TMOS instead of TEOS in a similar procedure at 423 K gave similar adsorption trends for water, methanol, and ethanol vapor (see Figure 8 S in the Supporting Information). As evidenced by the solid-state NMR spectrum, the local condensation network around the silicon atoms in the silica replica is similar to that of conventional silica materials, such as mesoporous silica materials, that are hydrophilic due to the surface silanol groups. However, the silica replica of the HKUST-1 is relatively hydrophobic as shown in the adsorption experiments. The difference in the hydrophobic/hydrophilic nature of conventional silica and the material reported herein most likely derives from the difference in the condensation environments of the silica precursors. Although the template MOF has a relatively high affinity for water vapor, as evidenced by water-adsorption experiments,[53] the pore space is mainly surrounded by organic ligands, which should cause the pore Chem. Eur. J. 2015, 21, 12148 – 12152
www.chemeurj.org
Amorphous silica materials with well-defined micropores have been synthesized by replication of a metal–organic framework (MOF). The obtained silica material had a micro–meso binary pore system and micropores with pore sizes that correspond to the pore-wall thickness of the template MOF. The silica replica is hydrophobic, as confirmed by adsorption analyses. This hydrophobicity is due to the unique condensation environment of the silica precursors in the template MOF. By adopting this synthetic method, it was possible to define not only the structural features, such as the pore size and pore-network connectivity, but also the pore-surface properties. The success of the MOF replication method with both titania and silica suggests that this method could be applicable to other inorganic materials and also possibly to other MOF templates.
Experimental Section Materials Tetramethoxysilane (TMOS), tetraethoxysilane (TEOS), and [Cu(NO3)2]·3 H2O were purchased from the Wako Pure Chemical Industry Co. and used as received. The organic ligand 1,3,5-benzenetricarboxylic acid was purchased from the Tokyo Chemical Industry Co. and was used as received.
Synthesis of the MOF template The MOF HKUST-1 was selected as the template material. The synthesis of the MOF was carried out based on a previously reported procedure.[54] In a typical synthesis, [Cu(NO3)2] (2.17 g, 9.0 mmol) was dissolved in ethanol (30 mL), and the solution was slowly dropped into a solution of 1,3,5-benzenetricarboxylic acid (1.26 g, 6.0 mmol) in ethanol (30 mL). The mixture was stirred for 30 min and then heated at 353 K for two days. Finally, a blue powder was obtained after filtration and washing with water and ethanol (Yield 56 %).
Synthesis of porous silica In typical synthetic procedures, HKUST-1 was heated at 423 K under vacuum to remove guest molecules. Subsequently, the liquid silica precursors TMOS and TEOS were injected directly over the powder and left to stand for one day under reduced pressure at ambient temperature. The powder was washed with ethanol and collected by filtration. The dried powder was dispersed in a solution of ethanol/water (1:1, v/v), placed in a Teflon-lined autoclave, and heated overnight at 423–473 K. After the hydrothermal treatment, the resulting powder was filtered. The HKUST-1/silica composites were suspended in 0.1 m HCl to etch the template MOF, and the silica replica materials were collected after washing with
12151
Ó 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Full Paper water and ethanol. In the case of hydrothermal treatment at 423 K, for example, 50–80 mg of silica powder were obtained from 1 g of the template MOF. This corresponds to the amount ideally obtained from the composite with pore filling of 45–60 %.
Characterization Powder X-ray diffraction (XRD) patterns were measured on a Rigaku RINT-2100S diffractometer with monochromated CuKa radiation. EDX spectrometry was measured on a PHILIPS EDAX-DX-4 apparatus. IR spectroscopic data were collected at room temperature on a FTIR JASCO FTIR-4200 spectrometer. The spectra were collected over n˜ = 4600–600 cm¢1 by averaging 128 scans at a maximum resolution of 4 cm¢1. SEM images of HKUST-1 and the HKUST-1/silica composites after hydrothermal treatment at 423 K were taken on a JEOL JSM-6510 machine at 15 kV, and the images of the silica replicas were taken on Zeiss FESEM LEO electron microscope at 1 kV. Thermogravimetric differential thermal analysis (TG-DTA) was measured on a Rigaku Thermo Plus 2 instrument at a heating rate of 10 K min¢1 under an air flow of 100 mL min¢1. Nitrogen adsorption isotherms were measured by using the automatic volumetric apparatus Belsorp-28SA (Bel Japan, Japan) at 77 K. Adsorption isotherms of water, methanol, and ethanol were measured by using the different volumetric apparatus Belsorp-18 (Bel Japan, Japan) at 303 K. The samples were pretreated under vacuum (< 10¢1 Pa) at 423 K for 2 h prior to the gas-adsorption measurements.
Acknowledgements This work was supported by JSPS KAKENHI Grant Number 26810120. A.S.H. and T.E.M. acknowledge support from the National Science Foundation, grant DMR-1306938. Keywords: adsorption · hydrophobic effect · microporous materials · metal–organic frameworks · silica [1] N. K. Raman, M. T. Anderson, C. J. Brinker, Chem. Mater. 1996, 8, 1682 – 1701. [2] A. Sayari, S. Hamoudi, Chem. Mater. 2001, 13, 3151 – 3168. [3] S.-W. Song, K. Hidajat, S. Kawai, Langmuir 2005, 21, 9568 – 9575. [4] F. Hoffmann, M. Cornelius, J. Morell, M. Frçba, Angew. Chem. Int. Ed. 2006, 45, 3216 – 3251; Angew. Chem. 2006, 118, 3290 – 3328. [5] J. L. Vivero-Escoto, I. I. Slowing, B. G. Trewyn, V. S.-Y. Lin, Small 2010, 6, 1952 – 1967. [6] J. S. Beck, J. C. Vartuli, W. J. Roth, M. E. Leonowicz, C. T. Kresga, K. D. Schmitt, C. T.-W. Chu, D. H. Olson, E. W. Sheppard, S. B. McCullen, J. B. Higgins, J. L. Schlenker, J. Am. Chem. Soc. 1992, 114, 10834 – 10843. [7] D. Zhao, J. Feng, Q. Huo, N. Me losh, G. H. Fredrickson, B. F. Chmelka, G. D. Stucky, Science 1998, 279, 548 – 552. [8] D. Brunel, Microporous Mesoporous Mater. 1999, 27, 329 – 344. [9] R. Ryoo, C. H. Ko, M. Kruk, V. Antochshuk, M. Jaroniec, J. Phys. Chem. B 2000, 104, 11465 – 11471. [10] A.-H. Lu, F. Schìth, Adv. Mater. 2006, 18, 1793 – 1805. [11] I. Nowak, M. Jaroniec, Top. Catal. 2008, 49, 193 – 203. [12] P. F. Fulvio, A. Vinu, M. Jaroniec, J. Phys. Chem. C 2009, 113, 13565 – 13573. [13] A. Ivanova, D. Fattakhova-Rohlfing, B. E. Kayaalp, J. Rathousky´, T. Bein, J. Am. Chem. Soc. 2014, 136, 5930 – 5937. [14] N. K. Chaudhari, S. Chaudhari, J.-S. Yu, ChemSusChem 2014, 7, 3102 – 3111. [15] O. D. Velev, T. A. Jede, R. F. Lobo, A. M. Lenhoff, Nature 1997, 389, 447 – 448. Chem. Eur. J. 2015, 21, 12148 – 12152
www.chemeurj.org
[16] A. A. Zakhidov, R. H. Baughman, Z. Iqbal, C. X. Cui, I. Khayrullin, S. O. Dantas, I. Marti, V. G. Ralchenko, Science 1998, 282, 897 – 901. [17] S. A. Johnson, P. J. Ollivier, T. E. Mallouk, Science 1999, 283, 963 – 965. [18] O. D. Velev, E. W. Kaler, Adv. Mater. 2000, 12, 531 – 534. [19] A. Stein, N. R. Denny, Chem. Mater. 2008, 20, 649 – 666. [20] B. Mandlmeier, J. M. Szeifert, D. Fattakhova-Rohlfing, H. Amenitsch, T. Bein, J. Am. Chem. Soc. 2011, 133, 17274 – 17282. [21] T. Kyotani, T. Nagai, S. Inoue, A. Tomita, Chem. Mater. 1997, 9, 609 – 615. [22] S. A. Johnson, E. S. Brigham, P. J. Ollivier, T. E. Mallouk, Chem. Mater. 1997, 9, 2448 – 2458. [23] J. Rodriguez-Mirasol, T. Cordero, L. R. Radovic, J. J. Rodriguez, Chem. Mater. 1998, 10, 550 – 558. [24] Z. Ma, T. Kyotani, A. Tomita, Chem. Commun. 2000, 2365 – 2366. [25] H. Nishihara, T. Kyotani, Adv. Mater. 2012, 24, 4473 – 4498. [26] S. R. Batten, R. Robson, Angew. Chem. Int. Ed. 1998, 37, 1460 – 1494; Angew. Chem. 1998, 110, 1558 – 1595. [27] P. J. Hagrman, D. Hagrman, J. Zubieta, Angew. Chem. Int. Ed. 1999, 38, 2638 – 2684; Angew. Chem. 1999, 111, 2798 – 2848. [28] B. Moulton, M. J. Zawarotko, Chem. Rev. 2001, 101, 1629 – 1658. [29] O. M. Yaghi, M. O’Keeffe, N. W. Ockwig, H. K. Chae, M. Eddaoudi, J. Kim, Nature 2003, 423, 705 – 714. [30] L. Carlucci, G. Ciani, D. M. Proserpio, Coord. Chem. Rev. 2003, 246, 247 – 289. [31] S. Kitagawa, R. Kitaura, S. Noro, Angew. Chem. Int. Ed. 2004, 43, 2334 – 2375; Angew. Chem. 2004, 116, 2388 – 2430. [32] G. F¦rey, Chem. Soc. Rev. 2008, 37, 191 – 214. [33] M. P. Suh, H. J. Park, T. K. Prasad, D.-W. Lim, Chem. Rev. 2012, 112, 782 – 835. [34] M. Yoon, R. Srirambalaja, K. Kim, Chem. Rev. 2012, 112, 1196 – 1231. [35] B. Liu, H. Shioyama, T. Akita, Q. Xu, J. Am. Chem. Soc. 2008, 130, 5390 – 5391. [36] T. Uemura, D. Hiramatsu, K. Yoshida, S. Isoda, S. Kitagawa, J. Am. Chem. Soc. 2008, 130, 9216 – 9217. [37] T. Uemura, Y. Kadowaki, N. Yanai, S. Kitagawa, Chem. Mater. 2009, 21, 4096 – 4098. [38] L. Radhakrishnan, J. Reboul, S. Furukawa, P. Srinivasu, S. Kitagawa, Y. Yamauchi, Chem. Mater. 2011, 23, 1225 – 1231. [39] M. Hu, J. Reboul, S. Furukawa, N. L. Torad, Q. Ji, P. Srinivasu, K. Ariga, S. Kitagawa, Y. Yamauchi, J. Am. Chem. Soc. 2012, 134, 2864 – 2867. [40] C. W. Abney, K. M. L. Taylor-Pashow, S. R. Russell, Y. Chen, R. Samantaray, J. V. Lockard, W. Lin, Chem. Mater. 2014, 26, 5231 – 5243. [41] H. B. Wu, B. Y. Xia, L. Yu, X.-Y. Yu, X. W. Lou, Nat. Commun. 2015, 6, 6512 – 6519. [42] S. S.-Y. Chui, S. M.-F. Lo, J. P. H. Charmant, A. G. Orpen, I. D. Williams, Science 1999, 283, 1148 – 1150. [43] A. S. Hall, A. Kondo, K. Maeda, T. E. Mallouk, J. Am. Chem. Soc. 2013, 135, 16276 – 16279. [44] W.-H. Zhang, J. Lu, B. Han, M. Li, J. Xiu, P. Ying, C. Li, Chem. Mater. 2002, 14, 3413 – 3421. [45] T. Classen, M. Lingenfelder, Y. Wang, R. Chopra, C. Virojanadara, U. Starke, G. Costantini, G. Fratesti, S. Fabris, S. Gironcoli, S. Baroni, S. Haq, R. Raval, K. Kern, J. Phys. Chem. A 2007, 111, 12589 – 12603. [46] Z. Luan, J. A. Fournier, Microporous Mesoporous Mater. 2005, 79, 235 – 240. [47] T.-J. Ha, H.-H. Park, S.-B. Jung, H. Ryu, B.-G. Yu, J. Colloid Interface Sci. 2008, 326, 186 – 190. [48] M. A. Mondragûn, V. M. CastaÇo, J. Garcia, C. A. T¦llez, Vib. Spectrosc. 1995, 9, 293 – 304. [49] E. P. Barrett, L. G. Joyner, P. P. Halenda, J. Am. Chem. Soc. 1951, 73, 373 – 380. [50] G. Horvth, K. Kawazoe, J. Chem. Eng. Jpn. 1983, 16, 470 – 475. [51] M. M. Dubinin, Chem. Rev. 1960, 60, 235 – 241. [52] S. Brunauer, P. H. Emmett, E. Teller, J. Am. Chem. Soc. 1938, 60, 309 – 319. [53] P. Kìsgens, M. Rose, I. Senkovska, H. Frçde, A. Henschel, S. Siegle, S. Kaskel, Microporous Mesoporous Mater. 2009, 120, 325 – 330. [54] A. Kondo, S. Takanashi, K. Maeda, J. Colloid Interface Sci. 2012, 384, 110 – 115. Received: April 24, 2015 Published online on July 14, 2015
12152
Ó 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Full Paper
& Silica Materials
A New Synthetic Route to Microporous Silica with Well-Defined Pores by Replication of a Metal–Organic Framework Atsushi Kondo,*[a] Anthony Shoji Hall,[b] Thomas E. Mallouk,[b] and Kazuyuki Maeda[a] Abstract: Microporous amorphous hydrophobic silica materials with well-defined pores were synthesized by replication of the metal–organic framework (MOF) [Cu3(1,3,5-benzenetricarboxylate)2] (HKUST-1). The silica replicas were obtained by using tetramethoxysilane or tetraethoxysilane as silica precursors and have a micro–meso binary pore system. The BET surface area, the micropore volume, and the mesopore volume of the silica replica, obtained by means of hydrothermal treatment at 423 K with tetraethoxysilane, are
Introduction Inorganic oxides are important materials not only because of their mechanical and chemical stability, but also because of their versatile functionality as adsorbents, catalysts and supports, and as ferroic, magnetic, and dielectric materials. Silica materials are of particular interest for applications in molecular adsorption, separation, and drug delivery, and these materials provide a useful platform on which to covalently anchor various functionalities.[1–5] The broad scope of silica chemistry is largely based on the facile formation of its network structure, which can take various micro- and mesoporous forms, and the robust chemistry that exists for its surface modification. Mesoporous silica represents a successful example of nanoscale structural control that is achieved by means of colloid chemistry and sol–gel chemistry.[6–9] In the assembly of mesoporous silica materials, surfactant species with different chain lengths result in different pore sizes and network connectivities. The resulting materials have adsorption and catalytic properties that strongly correlate to the pore dimensions and surface chemistry. However, soft template structures are often flexible and can be dependent on temperature, solvent, charge of the ions, and other parameters, which makes the prediction of the [a] Dr. A. Kondo, Prof. K. Maeda Department of Applied Chemistry, Graduate School of Engineering, Tokyo University of Agriculture and Technology 2-24-16 Naka-cho, Koganei, Tokyo 184-8588 (Japan) Fax: (+ 81) 42-388-7040 E-mail: [email protected] [b] Dr. A. S. Hall, Prof. T. E. Mallouk Department of Chemistry The Pennsylvania State University 104 Chemistry Building, University Park, PA 16802 (USA) Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/chem.201501599. Chem. Eur. J. 2015, 21, 12148 – 12152
620 m2g¢1, 0.18 mL g¢1, and 0.55 mL g¢1, respectively. Interestingly, the silica has micropores with a pore size of 0.55 nm that corresponds to the pore-wall thickness of the template MOF. The silica replica is hydrophobic, as confirmed by adsorption analyses, although the replica has a certain amount of silanol groups. This hydrophobicity is due to the unique condensation environment of the silica precursors in the template MOF.
resulting negative replica more difficult; therefore, other synthetic methodologies have been investigated to broaden the scope of silica chemistry. One promising route to oxide materials with well-defined pore networks is through the replication of ordered materials that act as hard templates.[10–14] For example, spherical particles are used in the colloidal-crystal templating process, which enables the synthesis of many different types of replica materials with three-dimensional porosity on length scales of tens of nanometers to microns.[15–20] The pore connectivity and dimensions of replica materials derive from the structural characteristics of the template. For example, zeolites can be utilized to fabricate microporous carbon materials of extremely high surface area, and the structure and porosity depends on the dimensionality of the original zeolite pore network.[21–25] Recently, metal–organic frameworks (MOFs) or porous coordination polymers, which are composed of metal ions and organic linkers, have emerged as promising materials for gas storage/separation, sensing, and catalysis.[26–34] Because MOFs have a variety of structures with different pore sizes, shapes, and network connectivities, these materials are good candidates as hard templates for fabricating unique materials with micro- and mesoporous structures.[35–41] Very recently, we demonstrated this idea by making microporous titania replicas of the coppercontaining cubic MOF [Cu3(1,3,5-benzenetricarboxylate)2] (HKUST-1).[42, 43] This MOF provides an ordered nanospace that can withstand the conditions of a hydrothermal reaction, but is soluble enough to be removed under very mild conditions. Herein, we explore the synthesis of silica replicas of this MOF. Although the pore size of the MOF is close to the size of the silica precursor, the precursor molecules infiltrate into the micropores and form a silica network in a hydrothermal reaction. The final silica product is not crystallographically ordered, but has well-defined micropores with a pore size that corre-
12148
Ó 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Full Paper sponds to the pore-wall thickness of the template material. Interestingly, adsorption experiments show that the resulting microporous silica materials are hydrophobic. The success of the MOF replication method with silica suggests that it may be possible to generalize the method to other oxides and perhaps to other MOF templates, thus leading to materials of versatile compositions and structures.
Results and Discussion XRD patterns of the HKUST-1 template and HKUST-1/silica composites obtained with tetraethoxysilane (TEOS) as the silica source after hydrothermal treatment are shown in Figure 1 a. The composites retain the structure of the HKUST-1 template up to 423 K. At higher hydrothermal-treatment temperatures (453 and 473 K), prominent diffraction lines at 36.4 and 42.38 2q appear in the XRD pattern, thus signaling the formation of Cu2O. Therefore, HKUST-1 can act as a hard template up to 423 K without decomposition to form silica composites. It is interesting that HKUST-1 retains its structure at higher temperatures (up to 473 K) when tetramethoxysilane (TMOS) was used instead as the silica precursor (see Figure 1 S in the Supporting Information), possibly because of the higher ceramic fraction of silica in TMOS relative to TEOS, which means the smaller contraction of TMOS molecules than TEOS through the condensation reaction to form the silica replica. Figure 1 b shows XRD patterns of the silica replicas obtained after the removal of the HKUST-1 template by HCl etching. The XRD patterns in Figure 1 b indicate that the silica replicas are amorphous and that the microcrystalline Cu2O is completely removed by the HCl etching. A broad amorphous halo peak was observed in the range of 20–308 for the sample treated at 423 K. This peak is weaker for the samples treated at higher temperatures, which may imply a decrease in short-range ordering of the
Figure 1. a) XRD patterns of the HKUST-1 template and the HKUST-1/silica composites obtained with TEOS as the silica source after hydrothermal treatment in the range of 353–473 K. The asterisk symbol (*) indicates diffraction peaks of Cu2O. b) XRD patterns of silica materials obtained after HKUST-1 removal. Chem. Eur. J. 2015, 21, 12148 – 12152
www.chemeurj.org
SiO4 tetrahedra. The temperature of the hydrothermal treatment was critical for the formation of silica networks in the pores of HKUST-1 because no solid was recovered after HCl etching when the composites made from TEOS were treated at 353 or 393 K. In contrast, the use of TMOS as a precursor gave solid silica products over a wide range of hydrothermal temperatures (i.e., 393–473 K). The silica replica obtained by hydrothermal treatment at 423 K by using TEOS was selected as a representative and was characterized. The energy-dispersive X-ray (EDX) spectrum of the sample predominantly shows the presence of silicon atoms and the absence of metal species, including copper, which is the metal ion in the HKUST-1 template (see Figure 2 S in the Supporting Information). The curve obtained by thermogravimetry (TG) of the etched material shows a gradual weight loss of 5 wt % in the temperature range 373–1273 K, thus indicating almost complete removal of organic moieties from the HKUST-1 template (see Figure 3 S in the Supporting Information). A solid-state 29Si cross-polarization magic-angle spinning (CP-MAS) NMR spectrum of the silica replica exhibited a peak at d = ¢100 ppm, which is assigned to the Q3 site with peaks at d = ¢91 and ¢111 ppm (Q2 and Q4 sites, respectively; Qn = Si(OT)n(OH)4¢n ; see Figure 4 S in the Supporting Information). Although it is not possible to conduct a quantitative analysis of the results obtained by NMR spectroscopic analysis because of the cross-polarization technique used, it is clear that the silica replica contains a considerable amount of Q3 units in addition to fully connected Q4 units and partially polymerized Q2 units. A mixture of silicon connectivity that ranges from Q2 to Q4 is also typically observed in amorphous porous silica materials, such as SBA-15,[44] thus indicating that the local silica network structure in the silica replica of HKUST-1 is similar to that of the amorphous silica materials. Figure 2 shows FTIR spectra of the etched material together with the HKUST1 template and the composite after hydrothermal treatment. The IR spectrum of the composite indicates the presence of 1,3,5-benzenetricarboxylate (BTC) ligands because of the strong absorption bands at n˜ = 1374 and 1647 cm¢1 , which are assignable to the symmetric and asymmetric stretching modes, respectively, of a carboxylate group.[45] The absence of absorption bands in the region n˜ = 1720–1700 cm¢1 suggests the absence of free carboxylic acid. Therefore, it is reasonable
Figure 2. FTIR spectra of the HKUST-1 template (upper), the composite after hydrothermal treatment (middle), and the silica replica obtained by HCl etching (bottom). The absorption bands of the carboxylate groups of the BTC ligand (+ +) and the asymmetric Si¢O bending mode of the SiO4 tetrahedra (*) are indicated.
12149
Ó 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Full Paper to suggest that coordination bonding between the carboxylate groups of the BTC ligands and copper(II) ions was retained throughout the hydrothermal treatment. The IR spectrum of the etched sample has no absorption bands assigned to BTC ligands, and an obvious broad peak in the range n˜ = 1000– 1200 cm¢1 is assigned to the asymmetric Si¢O bending mode of the SiO4 tertrahedra. In the spectrum of the etched material, a small broad absorption band in the region n˜ = 3600– 3100 cm¢1 is assignable to the stretching modes of the hydroxy groups, thus suggesting a certain amount of silanol groups (see Figure 5 S in the Supporting Information).[46, 47] In addition, the etched material shows no apparent absorption that corresponds to the C¢H vibration of alkyl chains in the range n˜ = 3000–2800 cm¢1, thus indicating the absence of alkoxy groups from the silica precursor molecules, which is consistent with the solid-state 13C MAS NMR results.[48] Many particles of the HKUST-1/silica composites retain the bipyramidal morphology of the parent MOF crystals, and no apparent silica particles were observed outside of HKUST-1 , thus suggesting that only negligible amounts of solid silica formed through the condensation reactions of the precursor molecules (see Figure 6 S in the Supporting Information). Figure 3 shows SEM images of the silica replica obtained by etching of the composite material. Although the etched silica materials do not perfectly retain the macroscale morphology of the MOF crystals, the particles have flat faces with sizes in the micrometer range, thus suggesting replication of the MOF template. A SEM image at high magnification (Figure 3 b) shows that the silica particles have fine pores in the range of mesopores (n = pore size; n = 2–50 nm) and macropores (n > 50 nm) formed from continuously connected silica networks.
Figure 3. a) A SEM image of the silica obtained by etching of the composite and b) an enlarged image of the boxed part of (a). Chem. Eur. J. 2015, 21, 12148 – 12152
www.chemeurj.org
Figure 4. a) Nitrogen adsorption–desorption isotherms of the silica replica at 77 K. Solid and open symbols represent adsorption and desorption branches, respectively. b) Pore-size distribution of the silica derived from the N2 adsorption isotherm at 77 K by using the HK method. In y axis, w and w0 means the pore volume and the micropore volume, respectively.
The adsorption properties of the silica replicas were examined to determine the pore-size distribution and pore volume. Figure 4 shows N2 adsorption–desorption isotherms of the silica material at 77 K. As expected from the SEM images, there is a hysteresis loop, which indicates the presence of mesopores. Because of the gradual increase over a wide range of relatively high pressures, the mesopore size is not uniform. This finding was confirmed by means of a Barrett–Joyner–Halenda (BJH) pore-size analysis[49] of the N2 adsorption isotherm (see Figure 7 S in the Supporting Information). A steep uptake at low relative pressure indicates the presence of micropores. The micropore-size distribution of the silica derived from the isotherm by using the Horvath–Kawazoe (HK) method[50] is shown in Figure 4 b. A sharp peak at 0.55 nm, which corresponds to the pore-wall thickness in the HKUST-1 template, is observed in the distribution function, although the function has a relatively broad distribution of micropore diameters. The micropore volume and isosteric heat of adsorption qst,f= 1/e evaluated by using the Dubinin–Radushkevich analysis[51] are 0.22 mL g¢1 and 11.0 kJ mol¢1, respectively. The BET surface area[52] of the silica was 620 m2 g¢1 and total pore volume was 0.77 mL g¢1. A mesopore volume of 0.55 mL g¢1 was obtained by subtracting the micropore volume from the total pore volume. Mesopores should be formed by incomplete filling of the HKUST-1 template with the silica component. Adsorption properties of other small molecules were also investigated. Figure 5 shows vapor-adsorption isotherms of water, methanol, ethanol, and tetrachloromethane on the silica
12150
Ó 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Full Paper surface of the silica replica to be hydrophobic. Copper(II) ions, which are arranged periodically in the template MOF, may have a catalytic influence to induce the condensation of the surface functional groups to form the hydrophobic pore surface. Thus, it should be noted that MOF templates have the potential to define not only the structural features, such as pore size and pore-network connectivity, but also the pore-surface properties.
Conclusion Figure 5. Adsorption isotherms of water (*), methanol (~), ethanol (&), and tetrachloromethane (^) on the obtained silica measured at 303 K. The solid and open symbols represent the adsorption and desorption branches, respectively.
replicas measured at 303 K. Relative to the methanol and ethanol isotherms, the water-adsorption isotherm shows a gentle increase in uptake over a wide relative pressure range with a smaller uptake amount. This behavior suggests that the material is relatively hydrophobic, which is consistent with the lack of silanol groups in the FTIR spectra. In contrast, the methanol- and ethanol-adsorption isotherms have steep uptakes in the low-pressure region and are very similar to each other. The amounts of adsorbed water, methanol, ethanol, and tetrachloromethane near P/P0 = 1 are 10.1, 16.3, 10.5, and 7.4 mmol g¢1, respectively. The ratio of the molecular volumes is 1:2.6:4.1:7.5 (water/methanol/ethanol/tetrachloromethane); therefore, the ratio of occupied volumes by the molecular species is 1.0:4.3:4.2:5.5 (water/methanol/ethanol/tetrachloromethane). This result clearly indicates that water molecules only partially occupy the pores of the silica, whereas methanol, ethanol, and tetrachloromethane occupy four to five times as much of the pore volume. The pore volumes evaluated for different molecular species are 0.18, 0.67, 0.61, and 0.71 mL g¢1, which are 23, 87, 79, and 92 % of the total pore volume estimated by N2 adsorption, for water, methanol, ethanol, and tetrachloromethane, respectively. The high filling percentage of tetrachloromethane directly indicates that the pore space in the silica replica is hydrophobic. The replicas made from TMOS instead of TEOS in a similar procedure at 423 K gave similar adsorption trends for water, methanol, and ethanol vapor (see Figure 8 S in the Supporting Information). As evidenced by the solid-state NMR spectrum, the local condensation network around the silicon atoms in the silica replica is similar to that of conventional silica materials, such as mesoporous silica materials, that are hydrophilic due to the surface silanol groups. However, the silica replica of the HKUST-1 is relatively hydrophobic as shown in the adsorption experiments. The difference in the hydrophobic/hydrophilic nature of conventional silica and the material reported herein most likely derives from the difference in the condensation environments of the silica precursors. Although the template MOF has a relatively high affinity for water vapor, as evidenced by water-adsorption experiments,[53] the pore space is mainly surrounded by organic ligands, which should cause the pore Chem. Eur. J. 2015, 21, 12148 – 12152
www.chemeurj.org
Amorphous silica materials with well-defined micropores have been synthesized by replication of a metal–organic framework (MOF). The obtained silica material had a micro–meso binary pore system and micropores with pore sizes that correspond to the pore-wall thickness of the template MOF. The silica replica is hydrophobic, as confirmed by adsorption analyses. This hydrophobicity is due to the unique condensation environment of the silica precursors in the template MOF. By adopting this synthetic method, it was possible to define not only the structural features, such as the pore size and pore-network connectivity, but also the pore-surface properties. The success of the MOF replication method with both titania and silica suggests that this method could be applicable to other inorganic materials and also possibly to other MOF templates.
Experimental Section Materials Tetramethoxysilane (TMOS), tetraethoxysilane (TEOS), and [Cu(NO3)2]·3 H2O were purchased from the Wako Pure Chemical Industry Co. and used as received. The organic ligand 1,3,5-benzenetricarboxylic acid was purchased from the Tokyo Chemical Industry Co. and was used as received.
Synthesis of the MOF template The MOF HKUST-1 was selected as the template material. The synthesis of the MOF was carried out based on a previously reported procedure.[54] In a typical synthesis, [Cu(NO3)2] (2.17 g, 9.0 mmol) was dissolved in ethanol (30 mL), and the solution was slowly dropped into a solution of 1,3,5-benzenetricarboxylic acid (1.26 g, 6.0 mmol) in ethanol (30 mL). The mixture was stirred for 30 min and then heated at 353 K for two days. Finally, a blue powder was obtained after filtration and washing with water and ethanol (Yield 56 %).
Synthesis of porous silica In typical synthetic procedures, HKUST-1 was heated at 423 K under vacuum to remove guest molecules. Subsequently, the liquid silica precursors TMOS and TEOS were injected directly over the powder and left to stand for one day under reduced pressure at ambient temperature. The powder was washed with ethanol and collected by filtration. The dried powder was dispersed in a solution of ethanol/water (1:1, v/v), placed in a Teflon-lined autoclave, and heated overnight at 423–473 K. After the hydrothermal treatment, the resulting powder was filtered. The HKUST-1/silica composites were suspended in 0.1 m HCl to etch the template MOF, and the silica replica materials were collected after washing with
12151
Ó 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Full Paper water and ethanol. In the case of hydrothermal treatment at 423 K, for example, 50–80 mg of silica powder were obtained from 1 g of the template MOF. This corresponds to the amount ideally obtained from the composite with pore filling of 45–60 %.
Characterization Powder X-ray diffraction (XRD) patterns were measured on a Rigaku RINT-2100S diffractometer with monochromated CuKa radiation. EDX spectrometry was measured on a PHILIPS EDAX-DX-4 apparatus. IR spectroscopic data were collected at room temperature on a FTIR JASCO FTIR-4200 spectrometer. The spectra were collected over n˜ = 4600–600 cm¢1 by averaging 128 scans at a maximum resolution of 4 cm¢1. SEM images of HKUST-1 and the HKUST-1/silica composites after hydrothermal treatment at 423 K were taken on a JEOL JSM-6510 machine at 15 kV, and the images of the silica replicas were taken on Zeiss FESEM LEO electron microscope at 1 kV. Thermogravimetric differential thermal analysis (TG-DTA) was measured on a Rigaku Thermo Plus 2 instrument at a heating rate of 10 K min¢1 under an air flow of 100 mL min¢1. Nitrogen adsorption isotherms were measured by using the automatic volumetric apparatus Belsorp-28SA (Bel Japan, Japan) at 77 K. Adsorption isotherms of water, methanol, and ethanol were measured by using the different volumetric apparatus Belsorp-18 (Bel Japan, Japan) at 303 K. The samples were pretreated under vacuum (< 10¢1 Pa) at 423 K for 2 h prior to the gas-adsorption measurements.
Acknowledgements This work was supported by JSPS KAKENHI Grant Number 26810120. A.S.H. and T.E.M. acknowledge support from the National Science Foundation, grant DMR-1306938. Keywords: adsorption · hydrophobic effect · microporous materials · metal–organic frameworks · silica [1] N. K. Raman, M. T. Anderson, C. J. Brinker, Chem. Mater. 1996, 8, 1682 – 1701. [2] A. Sayari, S. Hamoudi, Chem. Mater. 2001, 13, 3151 – 3168. [3] S.-W. Song, K. Hidajat, S. Kawai, Langmuir 2005, 21, 9568 – 9575. [4] F. Hoffmann, M. Cornelius, J. Morell, M. Frçba, Angew. Chem. Int. Ed. 2006, 45, 3216 – 3251; Angew. Chem. 2006, 118, 3290 – 3328. [5] J. L. Vivero-Escoto, I. I. Slowing, B. G. Trewyn, V. S.-Y. Lin, Small 2010, 6, 1952 – 1967. [6] J. S. Beck, J. C. Vartuli, W. J. Roth, M. E. Leonowicz, C. T. Kresga, K. D. Schmitt, C. T.-W. Chu, D. H. Olson, E. W. Sheppard, S. B. McCullen, J. B. Higgins, J. L. Schlenker, J. Am. Chem. Soc. 1992, 114, 10834 – 10843. [7] D. Zhao, J. Feng, Q. Huo, N. Me losh, G. H. Fredrickson, B. F. Chmelka, G. D. Stucky, Science 1998, 279, 548 – 552. [8] D. Brunel, Microporous Mesoporous Mater. 1999, 27, 329 – 344. [9] R. Ryoo, C. H. Ko, M. Kruk, V. Antochshuk, M. Jaroniec, J. Phys. Chem. B 2000, 104, 11465 – 11471. [10] A.-H. Lu, F. Schìth, Adv. Mater. 2006, 18, 1793 – 1805. [11] I. Nowak, M. Jaroniec, Top. Catal. 2008, 49, 193 – 203. [12] P. F. Fulvio, A. Vinu, M. Jaroniec, J. Phys. Chem. C 2009, 113, 13565 – 13573. [13] A. Ivanova, D. Fattakhova-Rohlfing, B. E. Kayaalp, J. Rathousky´, T. Bein, J. Am. Chem. Soc. 2014, 136, 5930 – 5937. [14] N. K. Chaudhari, S. Chaudhari, J.-S. Yu, ChemSusChem 2014, 7, 3102 – 3111. [15] O. D. Velev, T. A. Jede, R. F. Lobo, A. M. Lenhoff, Nature 1997, 389, 447 – 448. Chem. Eur. J. 2015, 21, 12148 – 12152
www.chemeurj.org
[16] A. A. Zakhidov, R. H. Baughman, Z. Iqbal, C. X. Cui, I. Khayrullin, S. O. Dantas, I. Marti, V. G. Ralchenko, Science 1998, 282, 897 – 901. [17] S. A. Johnson, P. J. Ollivier, T. E. Mallouk, Science 1999, 283, 963 – 965. [18] O. D. Velev, E. W. Kaler, Adv. Mater. 2000, 12, 531 – 534. [19] A. Stein, N. R. Denny, Chem. Mater. 2008, 20, 649 – 666. [20] B. Mandlmeier, J. M. Szeifert, D. Fattakhova-Rohlfing, H. Amenitsch, T. Bein, J. Am. Chem. Soc. 2011, 133, 17274 – 17282. [21] T. Kyotani, T. Nagai, S. Inoue, A. Tomita, Chem. Mater. 1997, 9, 609 – 615. [22] S. A. Johnson, E. S. Brigham, P. J. Ollivier, T. E. Mallouk, Chem. Mater. 1997, 9, 2448 – 2458. [23] J. Rodriguez-Mirasol, T. Cordero, L. R. Radovic, J. J. Rodriguez, Chem. Mater. 1998, 10, 550 – 558. [24] Z. Ma, T. Kyotani, A. Tomita, Chem. Commun. 2000, 2365 – 2366. [25] H. Nishihara, T. Kyotani, Adv. Mater. 2012, 24, 4473 – 4498. [26] S. R. Batten, R. Robson, Angew. Chem. Int. Ed. 1998, 37, 1460 – 1494; Angew. Chem. 1998, 110, 1558 – 1595. [27] P. J. Hagrman, D. Hagrman, J. Zubieta, Angew. Chem. Int. Ed. 1999, 38, 2638 – 2684; Angew. Chem. 1999, 111, 2798 – 2848. [28] B. Moulton, M. J. Zawarotko, Chem. Rev. 2001, 101, 1629 – 1658. [29] O. M. Yaghi, M. O’Keeffe, N. W. Ockwig, H. K. Chae, M. Eddaoudi, J. Kim, Nature 2003, 423, 705 – 714. [30] L. Carlucci, G. Ciani, D. M. Proserpio, Coord. Chem. Rev. 2003, 246, 247 – 289. [31] S. Kitagawa, R. Kitaura, S. Noro, Angew. Chem. Int. Ed. 2004, 43, 2334 – 2375; Angew. Chem. 2004, 116, 2388 – 2430. [32] G. F¦rey, Chem. Soc. Rev. 2008, 37, 191 – 214. [33] M. P. Suh, H. J. Park, T. K. Prasad, D.-W. Lim, Chem. Rev. 2012, 112, 782 – 835. [34] M. Yoon, R. Srirambalaja, K. Kim, Chem. Rev. 2012, 112, 1196 – 1231. [35] B. Liu, H. Shioyama, T. Akita, Q. Xu, J. Am. Chem. Soc. 2008, 130, 5390 – 5391. [36] T. Uemura, D. Hiramatsu, K. Yoshida, S. Isoda, S. Kitagawa, J. Am. Chem. Soc. 2008, 130, 9216 – 9217. [37] T. Uemura, Y. Kadowaki, N. Yanai, S. Kitagawa, Chem. Mater. 2009, 21, 4096 – 4098. [38] L. Radhakrishnan, J. Reboul, S. Furukawa, P. Srinivasu, S. Kitagawa, Y. Yamauchi, Chem. Mater. 2011, 23, 1225 – 1231. [39] M. Hu, J. Reboul, S. Furukawa, N. L. Torad, Q. Ji, P. Srinivasu, K. Ariga, S. Kitagawa, Y. Yamauchi, J. Am. Chem. Soc. 2012, 134, 2864 – 2867. [40] C. W. Abney, K. M. L. Taylor-Pashow, S. R. Russell, Y. Chen, R. Samantaray, J. V. Lockard, W. Lin, Chem. Mater. 2014, 26, 5231 – 5243. [41] H. B. Wu, B. Y. Xia, L. Yu, X.-Y. Yu, X. W. Lou, Nat. Commun. 2015, 6, 6512 – 6519. [42] S. S.-Y. Chui, S. M.-F. Lo, J. P. H. Charmant, A. G. Orpen, I. D. Williams, Science 1999, 283, 1148 – 1150. [43] A. S. Hall, A. Kondo, K. Maeda, T. E. Mallouk, J. Am. Chem. Soc. 2013, 135, 16276 – 16279. [44] W.-H. Zhang, J. Lu, B. Han, M. Li, J. Xiu, P. Ying, C. Li, Chem. Mater. 2002, 14, 3413 – 3421. [45] T. Classen, M. Lingenfelder, Y. Wang, R. Chopra, C. Virojanadara, U. Starke, G. Costantini, G. Fratesti, S. Fabris, S. Gironcoli, S. Baroni, S. Haq, R. Raval, K. Kern, J. Phys. Chem. A 2007, 111, 12589 – 12603. [46] Z. Luan, J. A. Fournier, Microporous Mesoporous Mater. 2005, 79, 235 – 240. [47] T.-J. Ha, H.-H. Park, S.-B. Jung, H. Ryu, B.-G. Yu, J. Colloid Interface Sci. 2008, 326, 186 – 190. [48] M. A. Mondragûn, V. M. CastaÇo, J. Garcia, C. A. T¦llez, Vib. Spectrosc. 1995, 9, 293 – 304. [49] E. P. Barrett, L. G. Joyner, P. P. Halenda, J. Am. Chem. Soc. 1951, 73, 373 – 380. [50] G. Horvth, K. Kawazoe, J. Chem. Eng. Jpn. 1983, 16, 470 – 475. [51] M. M. Dubinin, Chem. Rev. 1960, 60, 235 – 241. [52] S. Brunauer, P. H. Emmett, E. Teller, J. Am. Chem. Soc. 1938, 60, 309 – 319. [53] P. Kìsgens, M. Rose, I. Senkovska, H. Frçde, A. Henschel, S. Siegle, S. Kaskel, Microporous Mesoporous Mater. 2009, 120, 325 – 330. [54] A. Kondo, S. Takanashi, K. Maeda, J. Colloid Interface Sci. 2012, 384, 110 – 115. Received: April 24, 2015 Published online on July 14, 2015
12152
Ó 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
