Macroporous periodic mesoporous organosilicas with diethylbenzene bridging groups

Microporous and Mesoporous Materials 130 (2010) 180–188

Contents lists available at ScienceDirect

Microporous and Mesoporous Materials journal homepage: www.elsevier.com/locate/micromeso

Macroporous periodic mesoporous organosilicas with diethylbenzene bridging groups Brian J. Melde *, Brandy J. Johnson, Michael A. Dinderman, Jeffrey R. Deschamps Center for Bio/Molecular Science and Engineering, Naval Research Laboratory, Washington, DC 20375, United States

a r t i c l e

i n f o

Article history: Received 4 September 2009 Received in revised form 3 November 2009 Accepted 5 November 2009 Available online 10 November 2009 Keywords: Macroporous monolith Periodic mesoporous organosilica Bis(trimethoxysilylethyl)benzene Hierarchical material Polymerization-induced phase separation

a b s t r a c t Periodic mesoporous organosilicas (PMOs) with diethylbenzene bridging groups (DEB) in the pore walls are promising adsorbents for harmful compounds including nitroenergetics. The hierarchical macropore– mesopore structures developed here are expected to facilitate diffusion for such applications and to offer the potential for application in column formats. DEB-functionalized PMOs were synthesized with triblock copolymer PluronicÒP123 and the swelling additive mesitylene. Many of the materials were prepared through co-condensation of the DEB precursor bis(trimethoxysilylethyl)benzene with 1,2-bis(trimethoxysilyl)ethane (BTE) to improve mesostructure. Conditions were optimized to form micron-scale macropores by polymerization-induced phase separation. PMOs were also produced using only the DEB precursor. These materials exhibited relatively high surface areas (ca. 400 m2/g) and mesopore sizes (ca. 45 Å) as well as isolated macropores. As-synthesized materials were fabricated as centimeter-scale monoliths that became powders after multiple solvent refluxes. Soxhlet extraction could be used to preserve monolithic morphology while removing enough P123 to access mesopores. Nitrogen sorption, powder X-ray diffraction, SEM, TEM, and TGA were applied to characterize the PMOs. Published by Elsevier Inc.

1. Introduction Molecular sieves commonly referred to as periodic mesoporous organosilicas (PMOs) [1–4] have been intensely studied in the field of hybrid inorganic–organic materials [5–12]. They combine molecular-scale mixing of inorganic and organic components through bridged-polysilsesquioxanes [13] with the ordered nanoporosity of surfactant-templated mesoporous silicates such as the M41S [14,15] and SBA materials [16,17]. The organic ‘‘bridginggroups” between Si atoms have influence over the structure and functionality of the resulting materials. More rigid bridging groups, such as ethane, ethene, and benzene, generally yield mesopore systems of high order while the use of less rigid groups, such as diethylbenzene (DEB), can be detrimental to the formation of an ordered material. DEB bridging groups have been shown to offer greater affinity for organic targets than benzene bridging groups. This enhanced functionality is likely related to the flexibility that hampers the formation of organized DEB materials [18]. Several applications of DEB materials have been described including chemical adsorption, bioadsorption, and catalysis [19–24]. Selective or semi-selective adsorption of toxic organic

* Corresponding author. Address: 4555 Overlook Ave SW, Code 6900, Washington, DC 20375, United States. Tel.: +1 202 767 0591; fax: +1 202 767 9594. E-mail address: [email protected] (B.J. Melde). 1387-1811/$ - see front matter Published by Elsevier Inc. doi:10.1016/j.micromeso.2009.11.003

compounds utilizes both the high surface area resulting from mesoporosity and the pore wall functionality. For example, adsorption of phenolic compounds from aqueous solution was demonstrated using a DEB-bridged material prepared through condensation in the presence of micellar assemblies of cetyltrimethylammonium [19]. This material had a disordered mesopore system; however, the attempt demonstrated the potential of the DEB functionality in environmental applications. Additional ordered and semi-ordered materials containing DEB have been synthesized through co-condensing the DEB precursor with tetramethoxysilane (TMOS) [20] or 1, 2-bis(trimethoxysilyl)ethane (BTE) [21,22]. These approaches provided materials with improved mesoporosity and some DEB functionality. In an attempt to further enhance the functionality of the DEB materials, a process similar to molecular imprinting of polymers was developed. Non-ionic alkylene oxide surfactant BrijÒ76 was modified with 3,5-dinitrobenzoyl head groups and included as a percentage of the BrijÒ76 micelles. This resulted in a higher degree of selectivity in adsorption of nitroenergetic targets from aqueous solution by BTE–DEB co-condensates [22]. The approach was developed based on reports that illustrated the potential for imprinting benzene- or DEB-bridged organosilicas using a hydrophobic dinitrobenzene-alkyl molecule that was found to be difficult to solubilize with surfactant micelles [24,25]. An additional protocol has been described in which the synthesis of an ordered DEB PMO was aided through addition of mesitylene and a salt to the self-assembly process [21].

B.J. Melde et al. / Microporous and Mesoporous Materials 130 (2010) 180–188

Although the PMOs specified above had accessible mesopores, they were synthesized as powder precipitates. This can limit their application for water remediation, preconcentration, or separation due to a high pressure drop when packed in a column format. Fine particles may create more bed volumes in a column but also impede the flow of solvent. One approach to adapting PMOs for chromatography is to synthesize the mesoporous organosilicas in the form of spheres with defined diameters. These spheres can be packed with reduced back pressures [26,27]. An alternative approach is to incorporate hierarchical macropore (>50 nm)-mesopore (2–50 nm) structures in silicate materials. This approach is advantageous for any diffusion dependent application. The macropores can act as ‘‘superhighways” through a framework consisting of mesoporous organosilicate. Nakanishi and co-workers have extensively researched the use of polymerization (condensation)-induced phase separation in sol–gel reactions to design macroporous silicate monoliths [28]. Formation of siloxane networks acts as a ‘‘chemical cooling” process that leads to micrometer-scale separation between gel and fluid phases. When certain requirements are met, the morphology of phase separated domains follows the process of spinodal decomposition. These domains form a co-continuous structure that coarsens over time, eventually fragmenting. The sol–gel transition ‘‘freezes” the morphology at a particular stage of development. Depending on how early the onset of phase separation is relative to gelation, open pore spaces are left upon removal of the fluid phase. Polymerization-induced phase separation has been successfully combined with surfactant templating to synthesize macroporous silicate and organosilicate monoliths in which the macroframeworks contain ordered mesopore structures [29–32]. Hierarchical macroporous–mesoporous monoliths of ethanebridged PMOs [29] and benzene-bridged silicas [33,34] have been described. Here, we report on the development of hierarchical macroporous–mesoporous DEB-bridged organosilicates. As indicated above, DEB offers enhanced functionality over benzene and ethane bridging groups. The utility of the DEB materials in adsorption of nitroenergetic targets from aqueous solution makes the development of materials applicable to column formats desirable. Materials described here include hierarchical co-condensed BTE–DEB materials, hierarchical DEB materials, and mesoporous DEB materials into which additional functionality has been incorporated through the inclusion of pendant groups. 2. Experimental 2.1. Chemicals 1, 2-bis(trimethoxysilyl)ethane 96% (BTE), mesitylene (1,3,5trimethylbenzene or TMB), nitric acid 70%, and hydrochloric acid 37% were obtained from Sigma–Aldrich; bis(trimethoxysilylethyl)benzene (DEB, mixture of 1, 4 and 1, 3 isomers), 3-aminopropyltrimethoxysilane (APTS), and phenyltrimethoxysilane (PTS) were obtained from Gelest, Inc.; and ethanol 200 proof was obtained from the Warner–Graham Company. PluronicÒP123 was generously donated by BASF. All chemicals were used as received. Water was deionized to 18.2 MX-cm by a Millipore Milli Q UV-Plus water purification system. 2.2. Synthesis of ethane-diethylbenzene-silicas (ED) and diethylbenzene-silicas (D) 1.9 g P123 and a determined amount of TMB were dissolved in 0.1 M HNO3 with stirring at 60 °C; exact amounts are listed in Tables 1 and 3 . The stirring solution was allowed to cool to RT and a

181

silane mixture consisting of 0.00784 mol total bis silane (BTE + DEB or DEB only) was added drop-wise. The reaction mixture was stirred until homogeneous and then transferred to a culture tube which was sealed tightly and heated at 60 °C over night (approximately 18 h). A white gel formed during this period. The tube was unsealed and heated at 60 °C for 2 d, and then 80 °C for 2 d. The product in the form of a white monolith was refluxed three times in ethanol for at least 12 h to extract P123, a process that rendered the monolith a powder. Materials were often refluxed a fourth time in 1 M HCl in ethanol to ensure a more thorough removal of P123. The material was collected by suction filtration, rinsed with ethanol and water, and dried at 100 °C. 2.3. Synthesis of ethane-diethylbenzene-phenyl-silicas (EDP) 1.9 g P123 and a determined amount of TMB (see Table 2) were dissolved in 7.5 g 0.1 M HNO3 with stirring at 60 °C. The stirring solution was cooled to RT and the silane mixture consisting of 0.0157 mol total Si (DEB, BTE and PTS) was added drop-wise. For materials with amine functionality (EDPA), 1.9 g P123 and 0.55 g TMB were dissolved in 9.5 g 0.1 M HNO3. Silanes (DEB, BTE and PTS) were added in the same molar ratio with an addition of 0.05 g APTS (0.000279 mol). The reaction mixture was stirred until homogeneous and transferred to a culture tube which was sealed tightly and heated at 60 °C over night (18 h). The tube was unsealed and the white gel was heated at 60 °C for 2 d and then 80 °C for 2 d. P123 was extracted by refluxing the monolith three times in ethanol for at least 12 h. A powdered product was collected by suction filtration and was rinsed with ethanol, then rinsed with water, and dried at 100 °C. 2.4. Characterization Nitrogen adsorption experiments were performed on a Micromeritics ASAP 2010 porosimeter at 77 K (Micromeritics Instrument Corporation, Norcross, GA). Samples were degassed to 1 lm Hg at 100 °C prior to analysis. Surface area was determined by use of the Brunauer–Emmett–Teller (BET) method, pore size was calculated by the Barrett–Joyner–Halenda (BJM) method from the adsorption branch of the isotherm, and total pore volume was determined by the single point method at relative pressure (P/P0) 0.97. Thermogravimetric analysis was performed using a TA Instruments Hi-Res 2950 thermogravimetric analyzer under a nitrogen atmosphere; temperature was ramped 5 °C/min to 800 °C (TA Instruments, Inc., New Castle, DE). Powder X-ray diffraction patterns were collected at room temperature using Cu Ka radiation from a Brüker MICROSTAR-H X-ray generator operated at 40 kV and 30 mA equipped with a 3 m Radian collimator, and a Brüker Platinum-135 CCD area detector. A custom fabricated beamstop was mounted on the detector to allow data collection to approximately 0.4° 2h (approximately 210 Å) with a sample to detector distance of 30 cm. After unwarping the images the XRD2 plug-in was used to integrate the diffraction patterns from 0.5° to 8.4° 2h. Scanning electron microscopy (SEM) samples were mounted on SEM stubs using conducting carbon tape. Sputter coating with gold under argon was accomplished using an auto sputter coater (Cressington 108) for a duration of 60 s. SEMs were collected using a LEO 1455 SEM (Carl Zeiss SMT, Inc., Peabody, MA). Instrument settings were as follows: Tungsten filament, secondary electron detector, 20.00 kV beam voltage, 300 V collector bias, 30.00 mm aperture, 6 mm working distance. Transmission electron microscopy (TEM) samples were combined with absolute ethanol, sonicated for 15 min at room temperature, deposited onto a holey carbon grid (200 mesh copper, SPI, West Chester, PA), and viewed under an energy filtering transmission electron microscope (LIBRA 120 EFTEM, Carl Zeiss SMT, Peabody, MA) at 120 kV. TEM images

182

B.J. Melde et al. / Microporous and Mesoporous Materials 130 (2010) 180–188

Table 1 Nitrogen adsorption data for ethane-diethylbenzene-bridged silica materials.

a

Material

Ratio BTE:DEB (0.00784 mol silane)

Mass TMB (g) (1.9 g P123)

Mass 0.1 M HNO3 (g)

BET specific surface area (m2/g)

Single point total pore volume (cm3/g)

BJH adsorption pore size (Å)

ED1 ED2 ED3 ED4 D0 ED5 ED6 ED7 ED8 ED9 ED10 ED11 ED12 EDA1 ED13

50:50 50:50 50:50 25:75 0:100 50:50 50:50 50:50 50:50 50:50 50:50 50:50 50:50 50:50a 50:50

0.3 0.35 0.4 0.4 0.4 0.3 0.3 0.3 0.3 0.35 0.4 0.5 0.55 0.55 0.6

6.07 6.07 6.07 6.07 6.07 6.5 7.0 7.5 8.5 7.5 7.5 7.5 7.5 7.5 7.5

606 614 587 427 75 675 696 715 648 589 691 722 683 648 712

0.509 0.548 0.471 0.310 0.0549 0.599 0.633 0.653 0.644 0.527 0.632 0.632 0.667 0.587 0.749

44 49 43 35 29 49 49 49 55 46 49 49 55 48 65

APTS (0.05 g) included in the synthesis.

Table 2 Nitrogen adsorption data for phenyl-functionalized ethane-diethylbenzene-bridged silica materials with and without amine modification. Material

Ratio BTE:DEB:PTMS (0.0157 mol Si)

Mass TMB (g) (1.9 g P123)

Mass 0.1 M HNO3 (g)

Mass APTMS (g)

BET specific surface area (m2/g)

Single point total pore volume (cm3/g)

BJH adsorption pore size (Å)

EDP1 EDP2 EDP3 EDP4 EDP5 EDPA1 EDPA2 EDPA3 EDPA4

50:40:10 50:40:10 50:40:10 50:40:10 50:40:10 50:40:10 60:30:10 70:20:10 75:15:10

0.4 0.5 0.55 0.6 0.7 0.55 0.55 0.55 0.55

7.5 7.5 7.5 7.5 7.5 9.5 9.5 9.5 9.5

0 0 0 0 0 0.05 0.05 0.05 0.05

422 430 440 420 456 64 90 93 224

0.392 0.399 0.434 0.450 0.511 0.0848 0.121 0.124 0.279

43 43 43 48 57 46 46 46 46

Table 3 Nitrogen adsorption data for diethylbenzene-bridged silica materials. Material (0.00784 mol silane)

Mass TMB (g) (1.9 g P123)

Mass 0.1 M HNO3 (g)

BET specific surface area (m2/g)

Single point total pore volume (cm3/g)

BJH adsorption pore size (Å)

D1 D2 D3 D4 D5 D6 D7 D8

0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

7.5 7.5 7.5 7.5 7.5 7.5 7.5 7.5

375 368 455 419 426 386 414 365

0.256 0.254 0.414 0.391 0.411 0.345 0.419 0.305

32 32 45 45 44 44 49 35

were captured on a bottom-mounted, digital camera (KeenView, Olympus SIS, Montvale, NJ). 3. Results and discussion 3.1. Ethane-diethylbenzene-bridged silicas (ED) Based on previous success in the application of 50% bis(trimethoxysilylethyl)benzene (DEB) bridged mesoporous materials, a hierarchical material of similar composition was desired. This material was expected to offer the binding characteristics of the mesoporous material with improved performance in a column format. The procedure of Nakanishi et al. for synthesizing macroporous– mesoporous monoliths of ethane-bridged silica [29] was adapted to accommodate the incorporation of 50–75 mol% DEB. The amount of 0.1 M aqueous nitric acid was fixed at 6.07 g and materials were synthesized using 0.3, 0.35, and 0.4 g TMB. A tempera-

ture of 80 °C was effective for evaporation drying. After refluxing in ethanol to remove P123, nitrogen sorption isotherms of the products designated ED1–ED3 were type IV in shape with H2 hysteresis between the adsorption and desorption branches (Fig. 1a). The desorption branches closed with the adsorption isotherms ca. relative pressure 0.4. This is the lower limit of stability for a meniscus of capillary condensed nitrogen and, therefore, does not necessarily reflect the presence of pore obstructions or irregularities [35]. Pore size distributions for the materials were relatively narrow and indicative of uniform mesopores (Fig. S1). Porosity data for the ED products are listed in Table 1. Increasing the amount of TMB in the syntheses did not significantly alter the measured surface areas (ca. 600 m2/g) or pore volumes (0.47– 0.55 cm3/g) for the resulting materials. Although TMB is normally an effective micelle swelling agent applied to obtain larger mesopores, the data did not demonstrate a direct relationship between TMB concentration and pore size (Table 1). It is possible that p–p

B.J. Melde et al. / Microporous and Mesoporous Materials 130 (2010) 180–188

183

Fig. 1. Nitrogen sorption isotherms of ethane-diethylbenzene silica materials (a) ED1 shifted by 200, ED2 shifted by 75, ED3, ED4, and D0 (b) ED5, ED6 shifted by 125, ED7 shifted by 275, ED8 shifted by 450, and EDA1 shifted by 600; (c) ED9, ED10 shifted by 100, ED11 shifted by 250, ED12 shifted by 400, and ED13 shifted by 550 cm3/g STP; powder X-ray diffraction patterns of (d) ED1–ED4, D0; (e) ED5–ED8, EDA1 and (f) ED9–ED13.

interactions between phenylene bridges in DEB might counter the activity of TMB under some conditions. This could lead to penetration of the micelles with resultant lower pore sizes. More evidence of this possibility will be presented below. Overall, porosity values were lower than those of similarly synthesized materials consisting of 100% BTE. BTE materials were found to have surface areas of 700–800 m2/g, pore volumes greater than 1 cm3/g, and mesopore sizes ca. 75 Å (not shown). Powder X-ray diffraction patterns of ED1–ED3 (Fig. 1d) displayed a peak in the low angle region near 1° 2h which could be attributed to the (1 0 0) reflection common for mesoporous materials with hexagonal order. However, a weak increase in intensity was observed in the general region where resolved (1 1 0) and (2 0 0) reflections would be expected. This indicated that order did not encompass long ranges. The low angle XRD peak, produced by these materials, was notably absent in a nonporous 50 BTE:50 DEB material synthesized without TMB additive. Incorporating greater amounts of DEB in the materials was, as expected, detrimental to mesoporosity. ED4 was synthesized with 75% DEB and had a reduced mesopore size of 35 Å. The isotherm and XRD characteristics for ED4 were similar to those of ED1– ED3. A 29 Å pore size was measured for a product synthesized using 100% DEB (D0), but it yielded a very low pore volume of 0.0549 cm3/g and an isotherm that had more Type I character than those of the ED products. Thermogravimetric analysis (TGA) showed that P123 accounted for only approximately 5% of the weight of D0 indicating that the low porosity was due to a lack of mesostructure and not to entrapped surfactant. XRD patterns of ED4 and D0 with more incorporated DEB and smaller pore sizes showed reflections ca. 1.1° 2h that were less intense than those observed for the ED1, ED2, and ED3 products. SEM images of the ED1–ED3 materials showed significant differences in morphology with variations in TMB concentration (Fig. 2a and b). ED1 (0.3 g TMB) exhibited spherical cavities with some, though not extensive, interconnectivity. Such distinct pores were not present in ED2 and ED3 synthesized with increased amounts of TMB (0.35 and 0.4 g, respectively). These materials were roughly textured and appeared to be agglomerations of small

irregularly shaped particles. When the amount of DEB was increased to 75% to produce ED4, a macroporous morphology similar to that of ED1 was obtained (Fig. 2c). Based on the results presented above, the amount of TMB in a synthesis was fixed at 0.3 g and the amount of aqueous nitric acid was varied (6.5–8.5 g) for 50% DEB materials. Macropores of indistinct shape with some torturous connectivity resulted (ED5–ED8; Fig. 2d). None of these materials had the highly open co-continuous macropore morphologies that have been reported for 100% ethane-bridged silicates. ED5–ED8 were found to have improved mesoporosity characteristics as compared to ED1–ED3 with surface areas greater than 600 m2/g, pore volumes greater than 0.6 cm3/g, and larger pore sizes (up to 55 Å for ED8). XRD reflections (Fig. 1e) appeared at slightly lower 2h values for these materials. Nitrogen sorption isotherms (Fig. 1b) indicated that capillary condensation occurred at higher relative pressure (P/P0), ca. 0.55 in the adsorption branches. The nitrogen desorption branches tended to close with the adsorption branches at P/P0 > 0.4, with ED8 having a type H1 hysteresis shape in which the branches were parallel to each other. To further investigate optimization of the materials, the nitric acid amount was fixed at 7.5 g and the amount of TMB was varied. The resulting materials had higher mesoporosity values (Table 1) and enhanced macroporous structures. SEM images of ED10 with 0.4 g of TMB showed evidence of micron-scale spherical cavities not seen in ED3 synthesized with less acid (Fig. 3a). Inclusion of 0.5–0.6 g of TMB (ED11–ED13) resulted in larger mesopore sizes (P49 Å) and nitrogen sorption isotherms with type H1 hystereses (Fig. 1c). The SEM of ED11 displayed aggregates of cylindrical and curved particles with less evidence of distinct spherical voids (Fig. 3b). A slightly greater amount of TMB (0.55 g) resulted in connectivity that formed a strut-like framework with extensive macroporosity (ED12; Fig. 3c). This structure strongly resembled the co-continuous morphology that was reported for P123-templated ethane-bridged [29] and benzene-bridged silicas [33,34]. Inclusion of 0.6 g TMB (ED13) coarsened the struts and somewhat disrupted the continuity of the macropores (Fig. 3d); however,

184

B.J. Melde et al. / Microporous and Mesoporous Materials 130 (2010) 180–188

Fig. 2. Scanning electron micrographs of (a) ED1, (b) ED2, (c) ED4 and (d) ED6.

ED13 had the highest mesoporosity characteristics of the materials synthesized for this study with a surface area of 712 m2/g, total pore volume of 0.749 cm3/g, and a mesopore size of 65 Å. Transmission electron micrographs of ED12 and ED13 are shown in Fig. 4. Hexagonally ordered mesostructures are clearly

Fig. 3. Scanning electron micrographs of (a) ED10, (b) ED11, (c) ED12 and (d) ED13.

observable in these images and those of various other ED materials. Secondary XRD peaks that are the fingerprint of well-ordered mesoporous silicates such as MCM-41 and SBA-15 were not observed. ED13 diffracted X-rays more intensely than other materials according to its powder pattern, but (1 1 0) and (2 0 0) reflections were

B.J. Melde et al. / Microporous and Mesoporous Materials 130 (2010) 180–188

Fig. 4. Transmission electron micrographs of (a) ED12 and (b) ED13.

not resolved (Fig. 1f). This can be interpreted as an effect of these ED products having more irregularities in their mesostructures compared to previously reported mesoporous materials and may contribute to the lack of DEB PMOs reported despite the commercial availability of the silane precursor. TGA of materials with high DEB compositions was used to evaluate the efficiency of surfactant removal from the silicate frameworks. It was determined that measurable amounts of P123 remained after three reflux extractions in ethanol. An example is presented in Fig. 5a of a 50% DEB product (ED12) after three ethanol extractions and after a fourth extraction with acidic ethanol. A small mass loss occurring before 100 °C of <5% is common due to adsorbed solvent and/or moisture. P123 decomposes by 400 °C and accounted for a loss of 21 wt.%. This tends to indicate that some triblock copolymer chains were entangled in the pore walls, although nitrogen sorption data proved that there was significant mesoporosity with a surface area of 456 m2/g despite incomplete removal of surfactant. The measured surface area increased to 683 m2/g after a subsequent extraction with 1 M hydrochloric acid in ethanol, and TGA weight loss due to P123 was reduced to 7%. Calcination of the materials for surfactant removal was not possible because, as indicated by TGA, bridging groups start to decompose before 300 °C. Multiple refluxes resulted in fine particle sizes, as is apparent in the SEM images. Characterization by solid-state NMR spectroscopy would confirm the incorporation of DEB bridging groups in the final products, however, this technique was unavailable. Mesoporous powders previously reported by Li et al. in which DEB was co-condensed

185

Fig. 5. Thermogravimetric analysis of (a) 50 BTE:50 DEB material (ED12) after three reflux extractions with ethanol and a fourth reflux extraction with 1 M HCl in ethanol; (b) 100% BTE (P19E-40) and 100% DEB (D0) materials after extraction.

with other silanes or used as a sole precursor were characterized using 13C cross-polarization magic-angle spinning and 29Si magic-angle spinning NMR spectroscopy [20,21]. It was determined that diethylbenzene bridging groups were incorporated in the pore walls for those materials without observable Si–C bond cleavage. We have previously reported significant enhancement in binding capacities for nitroenergetic targets upon increasing the ratio of DEB to BTE in mesoporous precipitates prepared by co-condensation [22]. The materials described here were synthesized as gels under relatively gentle conditions, so intact DEB bridging groups were expected in the pore walls. Fig. 5b provides TGA evidence of surfactant-extracted D0 (100% DEB) and a 100% BTE analog of composition P19E-40 reported by Nakanishi et al. [29]. Based on the difference in masses of ethane and diethylbenzene bridging groups, a 100% BTE material is expected to undergo a weight loss that is approximately 38% of that observed for a 100% DEB material. Although direct comparison is complicated by adsorbed solvent and surfactant residue as noted above along with incomplete combustion under a nitrogen atmosphere, a difference in the materials was apparent (Fig. 5b). TGA of the 100% BTE product showed a mass loss in the temperature range 200–600 °C that was approximately 39% of that measured for the 100% DEB material. Macroporous–mesoporous silica monoliths have potential as columns for chromatographic separations [36]. All of the materials described here were originally obtained as cylindrical monoliths as

186

B.J. Melde et al. / Microporous and Mesoporous Materials 130 (2010) 180–188

a result of using culture tubes for reaction chambers. Refluxing in alcohol crumbled the monoliths into fine particles, but was convenient for removing P123 without damaging the organic bridging groups in the pore walls. Monolithic morphology could be retained by performing a Soxhlet extraction (photograph shown in Fig. S2). The process is time-consuming and less efficient than refluxing; however, enough P123 could be removed to take advantage of the mesostructure of the material. A monolith (composition ED12) that was Soxhlet extracted with 1 M HCl in ethanol for 5 d had a measured surface area of 246 m2/g and pore volume equal to 0.421 cm3/g. Extraction with supercritical carbon dioxide using an alcohol co-solvent may offer a more efficient alternative for removing P123 if monolithic morphology in the material is desired [37]. 3.2. Ethane-diethylbenzene-bridged amine and phenyl silicas (EDA, EDP, and EDPA) Hierarchical pore structures are expected to enhance the utility of diethylbenzene-bridged silicas for adsorbing compounds of interest. PMOs with both DEB and amine functionalities can simultaneously adsorb phenyl compounds and metal ions [18]. Surface amine groups can also be used to anchor molecules for sensing or catalytic applications. A product (EDA1) was prepared similarly to macroporous–mesoporous ED12 with the addition of 0.05 g of 3-aminopropyltrimethoxysilane (APTS) in the reaction sol. The slightly reduced porosity values of EDA1 were expected due to the disrupting effect of the APTS precursor (Table 1). The material

was mesoporous with a surface area of 648 m2/g and a narrow pore size distribution at 48 Å. The direct addition of APTS to the synthesis appeared to coarsen and disrupt the co-continuous macropore structure seen for ED12. TEM confirmed that the material maintained a mesostructure with hexagonal order. Post-synthesis grafting would be an alternative for functionalizing the surfaces of an ED material while maintaining its co-continuous macropore morphology. Motivation to design porous adsorbents targeting particular toxic and nitroenergetic compounds led to further modification of the sorbent structure through incorporation of pendant phenyl groups. Materials were synthesized for which 10 mol% of the silicon in an ED (subtracted from the DEB component) was replaced with phenyltrimethoxysilane (PTS). EDP materials prepared with Si molar ratio 50 BTE:40 DEB:10 PTS were mesoporous as indicated by their type IV nitrogen sorption isotherms with H2 hystereses (Fig. 6a). They featured reduced surface areas, pore volumes, and pore sizes (Table 2) compared to their ED counterparts. Again, this was not unexpected for materials having pendant phenyl groups on the mesopore surfaces and/or encapsulated in the mesopore walls. The corresponding XRD peaks in Fig. 6c appeared at higher 2h values (lower d-spacings) indicating, along with the smaller pore sizes, that PTS did not contribute as much to mesopore wall thickness as DEB did. EDP materials were further modified through inclusion of APTS in the reaction sols. Addition of 0.05 g of APTS required 2 g more aqueous nitric acid to obtain a homogeneous mixture and prevent immediate precipitation or gelation. The resultant EDPA materials yielded type IV nitrogen sorption isotherms and

Fig. 6. Nitrogen sorption isotherms for phenyl-functionalized ethane-diethylbenzene-bridged silica materials (a) EDP1, EDP2 shifted by 100, EDP3 shifted by 200, EDP4 shifted by 325, and EDP5 shifted by 450; phenyl and amine-functionalized ethane-diethylbenzene-bridged silica materials (b) EDPA1, EDPA2 shifted by 25, and EDPA3 and EDPA4 shifted by 75 cm3/g STP; powder X-ray diffraction patterns of (c) EDP1–EDP5 and (d) EDPA1, EDPA3 and EDPA4.

B.J. Melde et al. / Microporous and Mesoporous Materials 130 (2010) 180–188

187

Fig. 7. Nitrogen sorption isotherms of diethylbenzene-bridged silica materials (a) D1, D2 shifted by 50, D3 shifted by 75, and D4 shifted by 200; (b) D5, D6 shifted by 100, D7 shifted by 200, and D8 shifted by 300 cm3/g STP; powder X-ray diffraction patterns of (c) D1–D4 and (d) D5–D8; scanning electron micrographs of (e) D4 and (f) D7.

narrow pore size distributions with greatly reduced surface areas and pore volumes (Fig 6b, Table 2) as well as XRD peaks of lower intensity (Fig. 6d). Increasing the amount of structure enhancing BTE at the expense of DEB to a ratio 75 BTE:15 DEB:10 PTS resulted in a surface area of 224 m2/g for EDPA4. Each EDPA material had a measured BJH adsorption pore size of 46 Å, regardless of the BTE: DEB ratio (Fig. S3). Phase separation or segregation of multiple bridging groups incorporated in the pore walls of PMOs has been investigated by Yang and Sayari [38]. There may exist domains that are relatively rich in BTE and DEB in these multifunctional PMOs for which templated mesoporosity is more prevalent in the ethane-silicate partitions. SEM images showed that the EDP materials and EDPA1 consisted of aggregates of irregularly shaped micrometer-scale particles. A co-continuous morphology was not observed, and any macropores in the materials resulted from voids formed by the aggregated micro-particles. There was a notable contrast between EDP3 and ED12 products synthesized with 0.55 g TMB and 7.5 g aqueous nitric acid. Replacement of 10 mol% Si from DEB with PTS in EDP3 prevented the formation of interconnected macroporosity observed for ED12. Gelation occurred quickly in these syntheses and likely solidified the product before the development of a micro-phase separation. Only bulk morphology was seen for EDPA2–EDPA4. 3.3. Diethylbenzene-bridged silicas (D) Previous reports have indicated increasing binding capacity for materials with greater concentrations of DEB [18,22]. It has been shown that this increased DEB concentration comes at the expense of mesoscale order within the materials and often results in reduced selectivity in binding. Mesoporous materials were obtained using 100% DEB (0.00784 mol) with 7.5 g of aqueous nitric acid and 0.3–1.0 g of TMB as evidenced by their distinct type IV nitrogen sorption isotherms in Fig 7a and b. It is notable that some isotherms have hystereses that do not close at low relative pressure. This has been observed with other surfactant-templated DEB silicates [20–22]. Nitrogen sorption analysis was repeated twice on

the same sample of D2 to insure that the technique itself did not cause restructuring. No variation was observed upon repetition of the measurement, so the cause of this hysteresis remains unclear. There was a significant change in porosity characteristics as well as XRD peak intensities between samples prepared using 0.4 and 0.5 g TMB (Table 3, Fig. 7c and d and Fig. S4). It seems that a crucial amount of swelling additive was required to obtain pore sizes comparable to those of 50% DEB materials. Products D3 and D4 prepared with 0.5 and 0.6 g of TMB respectively had predictably lower measured porosity characteristics than their counterparts with 50 BTE:50 DEB ratios (ED11 and ED13). Despite the absence of structure enhancing BTE, the materials had surface areas greater than 400 m2/g, total pore volumes near 0.4 cm3/g, and a mesopore size of 45 Å. To our knowledge, these values are the highest yet reported for surfactant-templated mesostructured diethylbenzenebridged silicas. Li et al. employed KCl with TMB to aid the interaction of P123 surfactant with silicate in acidic solution [21]. It should be noted that the cited material was a precipitate while the syntheses described here used more concentrated P123 solutions (20 wt.% in water) and a higher molar ratio of P123 to DEB to form monoliths. This approach made the use of a salt unnecessary in mesostructure assembly. Mesitylene was required for these syntheses, but relatively high amounts of TMB did not improve surface area or pore volume. D8 prepared with 1 g TMB was less mesoporous than D3 and D4. This result provided additional evidence related to how interactions between mesitylene and DEB can counter the swelling effect. Those interactions may have also prevented the morphological evolution of cylindrical to spheroid micelles in templating a mesostructured cellular foam [39,40] in this study. This is in contrast to what has been obtained with other macro-mesoporous silicas and organosilicas [29,31]. Regardless of the amount of TMB added, isolated spherical macropores were observed in the materials by SEM (Fig. 7e and f). 4. Conclusions Mesoporous organosilicas with diethylbenzene bridging groups in their pore walls were synthesized with PluronicÒP123 and the

188

B.J. Melde et al. / Microporous and Mesoporous Materials 130 (2010) 180–188

aid of mesitylene. Materials possessing micron-scale macroporous morphologies as a result of polymerization-induced phase separation were desired. A co-continuous macropore structure was obtained in a product synthesized with a 50:50 ratio of 1,2bis(trimethoxysilyl)ethane (BTE) and bis(trimethoxysilylethyl)benzene (DEB). While it was possible to further functionalize the structure with pendant phenyl and amine groups through cocondensation of silanes, those materials did not exhibit extensive macroporosity. The synthesis of 100% DEB materials with relatively high, uniform mesoporosity was also demonstrated. Materials of this type have been difficult to produce in the past due to the length and flexibility of the DEB bridging group. These materials featured isolated spherical macropores in addition to the mesoporous structure. Current ongoing efforts seek to build on these results. The goal is to obtain co-continuous macropore structures that should fulfill the potential of mesoporous DEB-bridged silicates for adsorption and concentration of targets. DEB is useful for binding particular compounds, including nitroenergetics, and P123 can be modified to ‘‘imprint” the surfaces of DEB silicas for selective adsorption. This process is similar to that reported for mesoporous materials synthesized using a BrijÒ76 surfactant [22]. The application of DEB-bridged silicas for preconcentration and catalysis will be the subject of forthcoming papers. Acknowledgements This research was sponsored by the US Department of Defense Strategic Environmental Research and Development Program (SERDP; ER-1604) and the Defense Threat Reduction Agency (DTRA; BA08PRO015). The views expressed here are those of the authors and do not represent those of the US Navy, the US Department of Defense, or the US Government. Appendix A. Supplementary data Nitrogen adsorption pore size distributions and a photograph of monolithic materials are available at www.elsevier.com. Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.micromeso.2009.11.003. References [1] S. Inagaki, S. Guan, Y. Fukushima, T. Ohsuna, O. Terasaki, J. Am. Chem. Soc. 121 (1999) 9611–9614. [2] B.J. Melde, B.T. Holland, C.F. Blanford, A. Stein, Chem. Mater. 11 (1999) 3302– 3308.

[3] C. Yohsina-Ishii, T. Asefa, N. Coombs, M.J. MacLachlan, G.A. Ozin, Chem. Commun. (1999) 2539–2540. [4] T. Asefa, M.J. MacLachlan, N. Coombs, G.A. Ozin, Nature 402 (1999) 867–871. [5] A. Stein, B.J. Melde, R.C. Schroden, Adv. Mater. 12 (2000) 1403–1419. [6] A. Sayari, S. Hamoudi, Chem. Mater. 13 (2001) 3151–3168. [7] B. Hatton, K. Landskron, W. Whitnall, D. Perovic, G.A. Ozin, Acc. Chem. Res. 38 (2005) 305–312. [8] F. Hoffmann, M. Cornelius, J. Morell, M. Froba, Angew. Chem. Int. Ed. 45 (2006) 3216–3251. [9] M.P. Kapoor, S. Inagaki, Bull. Chem. Soc. Japan 79 (2006) 1463–1475. [10] Y. Wan, D.Q. Zhang, N. Hao, D.Y. Zhao, Int. J. Nanotech. 4 (2007) 66–99. [11] S. Fujita, S. Inagaki, Chem. Mater. 20 (2008) 891–908. [12] Y.P. Zhang, C.H. Zhou, X.J. Wang, T. Yang, Y.Z. Xu, Prog. Chem. 20 (2008) 33–41. [13] D.A. Loy, K.J. Shea, Chem. Rev. 95 (1995) 1431–1442. [14] C.T. Kresge, M.E. Leonowicz, W.J. Roth, J.C. Vartuli, J.S. Beck, Nature 359 (1992) 710–712. [15] J.S. Beck, J.C. Vartuli, W.J. Roth, M.E. Leonowicz, C.T. Kresge, 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. 114 (1992) 10834–10843. [16] Q.S. Huo, D.I. Margolese, G.D. Stucky, Chem. Mater. 8 (1996) 1147–1160. [17] D. Zhao, Q. Huo, J. Feng, B.F. Chmelka, G.D. Stucky, J. Am. Chem. Soc. 120 (1998) 6024–6036. [18] S. Jayasundera, M.C. Burleigh, M. Zeinali, M.S. Spector, J.B. Miller, W. Yan, S. Dai, M.A. Markowitz, J. Phys. Chem. B 109 (2005) 9198–9201. [19] M.C. Burleigh, M.A. Markowitz, M.S. Spector, B.P. Gaber, Environ. Sci. Technol. 36 (2002) 2515–2518. [20] C. Li, J. Yang, X. Shi, J. Liu, Q. Yang, Micropor. Mesopor. Mater. 98 (2007) 220– 226. [21] C. Li, J. Liu, X. Shi, J. Yang, Q. Yang, J. Phys. Chem. C 111 (2007) 10948–10954. [22] B.J. Johnson, B.J. Melde, P.T. Charles, D.C. Cardona, M.A. Dinderman, A.P. Malanoski, S.B. Qadri, Langmuir 24 (2008) 9024–9029. [23] B. Johnson-White, M. Zeinali, A.P. Malanoski, M.A. Dinderman, Catal. Commun. 8 (2007) 1052–1056. [24] S.A. Trammell, M. Zeinali, B.J. Melde, P.T. Charles, F.L. Velez, M.A. Dinderman, A. Kusterbeck, M.A. Markowitz, Anal. Chem. 80 (2008) 4627–4633. [25] B. Johnson-White, M. Zeinali, K.M. Shaffer, J.C.H. Patterson, P.T. Charles, M.A. Markowitz, Biosens. Bioelectron. 22 (2007) 1154–1162. [26] D.J. Kim, J.-S. Chung, W.-S. Ahn, G.-W. Kang, W.-J. Cheong, Chem. Lett. 33 (2004) 422–423. [27] V. Rebbin, R. Schmidt, M. Fröba, Angew. Chem. Int. Ed. 45 (2006) 5210–5214. [28] K. Nakanishi, Porous Mater. 4 (1997) 67–112. [29] K. Nakanishi, Y. Kobayashi, T. Amatani, K. Hirao, T. Kodaira, Chem. Mater. 16 (2004) 3652–3658. [30] K. Nakanishi, K. Kanamori, J. Mater. Chem. 15 (2005) 3776–3786. [31] T. Amatani, K. Nakanishi, K. Hirao, T. Kodaira, Chem. Mater. 17 (2005) 2114– 2119. [32] K. Nakanishi, T. Amatani, S. Yano, T. Kodaira, Chem. Mater. 20 (2008) 1108– 1115. [33] D. Brandhuber, H. Peterlik, N. Huesing, Small 2 (2006) 503–506. [34] H. Zhong, G. Zhu, J. Yang, P. Wang, Q. Yang, Micropor. Mesopor. Mater. 100 (2007) 259–267. [35] M. Kruk, M. Jaroniec, Chem. Mater. 13 (2001) 3169–3183. [36] H. Minakuchi, K. Nakanishi, N. Soga, N. Ishizuka, N. Tanaka, Anal. Chem. 68 (1996) 3498–3501. [37] R.v. Grieken, G. Calleja, G.D. Stucky, J.A. Melero, R.A. Garcia, J. Iglesias, Langmuir 19 (2003) 3966–3973. [38] Y. Yang, A. Sayari, Chem. Mater. 20 (2008) 2980–2984. [39] P. Schmidt-Winkel, C.J. Glinka, G.D. Stucky, Langmuir 16 (2000) 356–361. [40] P. Schmidt-Winkel, J.W.W. Lukens, P. Yang, D.I. Margolese, J.S. Lettow, J.Y. Ying, G.D. Stucky, Chem. Mater. 12 (2000) 686–696.