- Email: [email protected]
Tunable hierarchical porous silica materials using hydrothermal sedimentation-aggregation technique
Microporous and Mesoporous Materials 208 (2015) 140e151
Contents lists available at ScienceDirect
Microporous and Mesoporous Materials journal homepage: www.elsevier.com/locate/micromeso
Tunable hierarchical porous silica materials using hydrothermal sedimentation-aggregation technique Abdelali Zaki a, Jing Xu a, Gregory Stoclet b, Sandra Casale c, Jean-Philippe Dacquin a, *, Pascal Granger a Unit e de Catalyse et Chimie du Solide (UCCS), CNRS UMR 8181, Universit e Lille I, Sciences et Technologies, 59655 Villeneuve d'Ascq, France Unit e Mat eriaux et Transformations (UMET), CNRS UMR 8207, Universit e Lille I, Sciences et Technologies, 59655 Villeneuve d'Ascq, France c Laboratoire de R eactivit e de Surface-CNRS UMR 7197, Universit e Pierre et Marie Curie, France a
b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 6 June 2014 Received in revised form 21 October 2014 Accepted 26 January 2015 Available online 3 February 2015
Herein, a simple and efficient synthesis based on a dual templating approach allow the preparation of tunable macroporous mesoporous silica materials. Macropores incorporation has been obtained using polymer spheres of well-defined size and their subsequent removal by thermal treatment allowed the independent control of the macropore size entrance between 200 nm and 50 nm. The addition of the block copolymer drives the formation of a second mesostructured skeleton, with a pore size centered around 4 nm, throughout the material framework. By a fine tuning of the solegel synthesis parameters (Polymer:TEOS ratio, aging conditions), we succeeded in guiding the SBA-15 rod-like morphology having randomly packed macropores to a derived SBA-15 with a homogeneous honeycomb macrostructure. Moreover, large mesoporous windows are generated between adjacent macropores. Hence, this simple one-pot synthesis approach, allowing scaling-up, offers fine tuning porosity at the macropore scale. © 2015 Elsevier Inc. All rights reserved.
Keywords: Hierarchical porous silicas SBA-15 Polymer spheres Hydrothermal sedimentaion-aggregation technique
1. Introduction Porous silica materials are widely used as supports for heterogeneous catalysis and separation applications since they offer various pore morphologies and high specific surface areas [1e5]. Moreover, with the emergence of alternative sources of energy, fuels and raw materials, the need for hierarchical architectures to prevent slow in-pore diffusion and poor atom efficiency of the reaction is highly desirable [6,7]. As a consequence, extensive researches on the design of heterogeneous solid supports have been developed since the end of the nineties [8e10] for maximizing the catalytic activity and reaching more sustainable technologies. In this line, templating strategies i.e. cooperative self-assembly [11], hard templates [12], biotemplates [13], phase separation [14] provide various possibilities for the synthesis of novel ordered mesostructures of tailored porosity and morphology. Various hierarchically structured silica materials can be obtained from the appropriate combination of such templating agents, including spray dried powders [15] and monoliths [16e18]. In these last
* Corresponding author. Tel.: þ33 (0)320434795. E-mail address: [email protected] (J.-P. Dacquin). http://dx.doi.org/10.1016/j.micromeso.2015.01.041 1387-1811/© 2015 Elsevier Inc. All rights reserved.
cases, surfactants such as cetylammonium salts or block copolymers can be added to induce a trimodal pore structure with micrometer-sized macropores. Dual-templating approach is as well an established technique and offers well-defined structural macroporosity and mesoporosity. Most often polymer spheres [19] exhibiting diameters of several hundred nanometers are used, and successful preparation of the macrostructured skeleton is crucially depending on the synthetic procedure usually based on the hard template pre-arrangement before infiltration by the surfactant-containing precursor [12,20]. This approach leads to the synthesis of hierarchical silica materials with an interconnected macropore network and mesopore network known as threedimensionally ordered macroporous materials with mesoporous walls (3 DOM/m materials) [21]. Research groups have also developed similar routes using polystyrene (PS) or polymethylmethacrylate (PMMA) spheres to produce 3 DOM/m transition metal oxides [22,23]. However, while the generation of smart and highly organized 3DOM structures can be obtained following this synthetic procedure, there is still restricted applications due to depth limitation in scaling-up [5]. Nevertheless, Ihm et al. proposed an alternative synthesis allowing feasible scaling-up with respect to the above described technique by using a solegel deposition of the silicate precursor on the polymer spheres
A. Zaki et al. / Microporous and Mesoporous Materials 208 (2015) 140e151
following hydrothermal sedimentation-aggregation technique [24] to generate a macroporous mesoporous MCM-41 [25]. This one-pot synthesis leads to a well-ordered macroporous skeleton with a mesoporous phase located between the walls of the macropores and having the same structure and arrangement of MCM-41. However, the direct application of this one pot synthesis to SBA15 materials, possessing better hydrothermal stability, led to a limited ordering of the macropores [26]. It has been found that the difference in the macrostructures between MCM-41 and SBA-15 originated from the difference in textural morphology of the parent mesoporous MCM-41 and SBA-15. Nevertheless, the incorporation of macropores (300 nm) in the mesoporous framework of the SBA-15 (5 nm) has been found to boost the catalytic activity for various relevant catalytic reactions [27e29]. Thus, the application of such macroporous mesoporous silica materials derived from SBA-15 demonstrated that an additional macroporous network could alleviate mass-transport limitations encountered in liquidphase reactions. Unfortunately, all those works [24e30] are reporting hierarchical porous silica materials with only one and large macropore size (300 nm). To our knowledge, hierarchical siliceous materials with sub<100 nm macropore size have been rarely reported in literature and are characterized by poorly ordered macrostructure with a broad macropore size distribution [31]. Herein, we report in a detailed study a simple and efficient one-pot synthesis allowing the preparation of hierarchical porous silica materials with sub-100 nm macropore size and bimodal mesoporosity. Our efforts were first devoted to the decrease of the polymer microsphere size (from 200 nm to 65 nm) and the research of right conditions to succeed in guiding the SBA-15 rod-like morphology having randomly packed macropores to a derived SBA15 having a well-organized honeycomb macrostructure with a mesoporous phase. Hence, improved textural properties have been found and could be highly desirable for liquid phase catalysis or adsorption of practical as well as fundamental interest. 2. Experimental section 2.1. Materials 2.1.1. Synthesis protocol of large polymer spheres (200 nm) Polystyrene spheres (PS) with a diameter size of 200 nm were prepared by an emulsifier-free emulsion polymerization technique and carried out in a 500 ml three neck bottom flask [24]. Typically, styrene (St) and divinylbenzene (DVB) are washed three times by an aqueous solution of NaOH (0.1 M) and then three times by distilled water in order to eliminate polymerization inhibitors. Then, 240 mL of distilled water is introduced in a 500 mL batch reactor. Before adding reagents, the aqueous solution is flushed under argon during at least 30 min in order to eliminate dissolved oxygen from the reactional medium. Then, 25.00 mL of St (monomer) and 4.75 mL of DVB (cross linker agent) are introduced in the reactor that is placed in a silicon oil bath under magnetic stirring and heating up to a fixed temperature (between 60 C and 90 C). After temperature stabilization, potassium persulfate (KPS) pre-dissolved in 20 mL of distilled water was injected to the reactional medium to initiate polymerization. The aqueous suspension of PS spheres was recovered after reaction by centrifugation or filtration and washed out with distilled water and dried at 80 C for 24 h. 2.1.2. Synthesis protocol of small polymer spheres (65 nm < PMMA < 100 nm) Polymethylmethacrylate spheres (PMMA) of diameter sizes between 65 nm and 100 nm were prepared by the surfactantassisted emulsion anionic polymerization technique [30] and carried out in a 500 mL three neck bottom flask. Typically, 330 mL of
141
distilled water is used as solvent and poured in a 500 mL three neck round bottom flask. Sodium dodecyl sulfate (SDS) surfactant and methyl methacrylate (MMA) monomers are added to the aqueous solution following 30 min of argon bubbling. Vigorous stirring is generally applied in order to have a good dispersion of MMA droplets and SDS. Then, the reactional medium is heated in a silicon oil bath at a desired temperature. After temperature stabilization, KPS pre-dissolved in 20 mL of distilled water was injected to the reactional medium to initiate polymerization. Argon bubbling is maintained during the experiment in order to limit undesired reactions which inhibit the polymer nodules growth. After 2 h reaction, the aqueous suspension of PMMA spheres is recovered by centrifugation and washed out with distilled water and dried at 80 C for 24 h. 2.1.3. Synthesis protocol of hierarchical porous silica materials (HS) Siliceous materials were synthesized following a similar dualtemplating route reported by Ihm et al. [26] Typically, 3.0 g of Pluronic P123 triblock copolymer was dissolved in 22.50 mL of water and 75.00 mL of 2 M HCl solution while stirring at 35 C. The required amount of polymer beads were then added to the solution and stirred for 1 h. Once a good dispersion of polymer beads was achieved, 6.90 mL of TEOS was added to the solution at 35 C and maintained for 24 h while stirring. The mixture was then transferred to the oven (T ¼ 80 C or 100 C) and maintained in a sealed polypropylene bottle under autogeneous pressure for a period of time ranging from 24 h to 96 h under static conditions. The solid product obtained is filtered, washed 3 times with deionised water and calcined in air at 550 C (ramp rate of 0.5 C min1) for 6 h. The final solids were labeled HS-X, where X ¼ polymer bead size. 2.1.4. Synthesis protocol of SBA-15 Pure mesoporous SBA-15 silica was synthesized following the approach of Zhao et al. [3] Typically, 3.0 g of Pluronic P123 triblock copolymer was dissolved in 22.50 mL of water and 75 mL of 2 M HCl solution while stirring at 35 C. 6.90 mL of TEOS (TEOS:P123 M ratio of 40) were added to the solution, which was maintained at 35 C for 24 h while stirring. The mixture was then aged at 80 C for 24 h and the solid product filtered, washed 3 times with deionised water, and calcined in air at 550 C (ramp rate of 0.5 C min1) for 6 h. 2.2. Characterisation techniques The average hydrodynamic diameters of PS and PMMA spheres were determined by dynamic light scattering (DLS, Nano ZS, Malvern Instruments). DLS was applied with an angle of 173 by using HeeNe laser (4 mW) operated at 633 nm. Average hydrodynamic diameters of nanoparticles were measured in aqueous solutions. Thermo-gravimetric analysis (TGA) was performed to study the thermal decomposition of polymer spheres. A SDT2960 TA instrument was used. 20 mg of sample were placed in an alumina crucible with a volume of 100 ml. The thermal analyzes were performed under air. The sample heated with a ramp in temperature of 5 C min1, from 25 to 700 C. Textural properties were carried out on an automated gas adsorption analyzer (Tristar 3020). Nitrogen sorption isotherms of the calcined samples were measured after outgassing at 200 C in vacuum for 3 h. The multipoint surface area was evaluated with the BrunauereEmmetteTeller (BET) method over the range P/P0 ¼ 0.075e0.35 and pore size distribution was obtained using BarretteJoynereHalenda (BJH) model applied to the desorption isotherm branch. Total pore volume was determined from the volume adsorbed at P/P0 ¼ 0.98. Transmission Electron micrographs were collected on a JEOL JEM 2011 (LaB6) operating at 200 kV while scanning emission
142
A. Zaki et al. / Microporous and Mesoporous Materials 208 (2015) 140e151
Table 1 Size controlled synthesis of anionic polymer nanospheres Sample
Monomer/mL
KPS/g
SDS/g
Stirring/rpm
Time reaction/h
Temperature/ C
dhyd/nm
dp/nm
PS-200 PS-145 PMMA-100 PMMA-65
25 24 24 24
0.083 0.187 0.187 0.187
/ 0.476 0.117 0.238
200 450 450 450
5 5 2 2
80 80 80 80
225 150 130 82
200 145 100 65
microscopy (Hitachi SU-70, SEM-FEG) was used for microstructure analysis of siliceous materials. Low angle X-Ray Diffraction (XRD) measurements was performed using a D8 advanced X-Ray diffractometer (Bruker AXS) equipped with a Cu-Ka (l ¼ 0.154 nm) radiation. Low angle XRD patterns were acquired from 0.3 to 6.0 with a 0.02 steps at 10 s per point. SAXS experiments were carried out using a Genix microsource (Xenocs, France) operated at 50 kV and 1 mA. The Cu-Ka radiation (l ¼ 0.154 nm) was selected and focused by means of FOX2D mirrors. The sample to detector distance was varied between 0.5 and 2 m. Silver Behenate was used as calibrant for the precise determination of the sample to detector distance. The 2D-SAXS patterns were recorded on a Photonic Science VHR CCD camera. Corrections were applied for background scattering, geometry and intensity distortions of the detector. Integrated intensity profiles I ¼ f(q), where q is the scattering vector (q¼(4p/l sin(q)/2)), were computed using the Fit2D software. Fitting of the scattered intensities I ¼ f(q) was performed using the Irena Package [32].
Fig. 1. Scanning electron images of (a) PS-200 and (b) PM-65 polymer spheres.
± ± ± ±
24 9 10 7
Yield/% 57 35 74 84
3. Results and discussions 3.1. Macrotemplate characterization Prior to the synthesis of the siliceous architectures, polymer spheres have been prepared using the emulsion polymerization technique. Large PS spheres have been successfully obtained following the surfactant-free emulsion polymerization which is one of the most common methods used for the production of polymers. Besides, by adding acetone surfactant within the mixture, PS sphere size can easily fall down according to literature [33]. However, only poor yields have been obtained in this study using St as monomer (Table 1). In order to reach lower sphere size while keeping a significant synthesis yield, the SDS surfactant assisted emulsion technique has been used to produce small PMMA nanospheres [30]. Results gathered in Table 1 and Fig. 1 show that nearly monodispersed spherical polymer spheres have been prepared with yields >50% and controllable size between 200 nm and 65 nm without altering the morphological properties. Thermal behavior of the macrotemplates (PMMA and PS) has been examined by thermogravimetric analysis (TGA) under air. Fig. 2
Fig. 2. TGA thermograms of weight loss and derivative weight loss as a function of temperature. (a) PM-65 and (b) PS-200 polymer spheres.
A. Zaki et al. / Microporous and Mesoporous Materials 208 (2015) 140e151
143
Fig. 3. Schematic illustration of the preparation of hierarchical porous silica by dual templating route.
Fig. 4. SEM micrographs of SBA-15 reference (a) and selected hierarchical porous silica materials with different Polymer:TEOS ratios (b) HS-200-3.5, (c) HS-65-1.5 and (d) HS-653.5.
Fig. 5. N2 adsorption desorption isotherms (A) and pore size distribution (B) of SBA-15 and macroporous mesoporous solids with different P/TEOS ratio (a) SBA-15, (b) HS-200-1.5, (c) HS-200-2.5 and (d) HS-200-3.5.
144
A. Zaki et al. / Microporous and Mesoporous Materials 208 (2015) 140e151
shows the sample weight loss and respective derivative weight loss as a function of temperature for PS and PMMA spheres (Fig. 2). As earlier evidenced in literature, the onset of the thermal decomposition of PS beads (287 C) is shifted toward the higher temperature range than that of PMMA beads (261 C), which confirms the enhancement of thermal stability of PS template. However, while the decomposition of PMMA proceeds in one step with a complete combustion of the organic material achieved at 350 C, the decomposition of PS spheres involves a two oxidation step process [34]. As a consequence, total weight loss of PS spheres is only achieved at 515 C. Based on these observations, the calcination temperature of hierarchical porous materials was set at 550 C.
3.2. Synthesis of tunable macroporous mesoporous silica materials Our dual-templating method involves the direct insertion of the macrotemplate in the solution following the addition of the TEOS and block-copolymer (Fig. 3). With a pH below 2, the solution containing the prehydrolysed precursor subsequently impregnate the anionic polymer spheres during one day and silica condensation reaction is achieved following hydrothermal treatment, resulting in an intermediate composite nanospheres/P123-silica structure. Hybrid materials were calcined in order to remove the organic porogens and to obtain the final siliceous solids. During this
Fig. 7. Low angle X-ray diffraction of SBA-15 reference and HS solids with optimised P:TEOS ratio. (a) SBA-15, (b) HS-65-3.5, (c) HS-100-3.5, (d) HS-200-3.5.
Fig. 6. N2 adsorption desorption isotherms (A) and pore size distribution (B) of macroporous mesoporous solids with different P/TEOS ratio. (a) HS-65-1.5 (b) HS-65-3.5, (c) HS-654.5, (d) HS-100-1.5, (e) HS-100-3.5.
A. Zaki et al. / Microporous and Mesoporous Materials 208 (2015) 140e151
145
Fig. 8. SEM micrographs of porous samples (A) HS-200-96h aged at 80 C; (B) HS-100-96h aged at 80 C and (C) HS-65-96h aged at 80 C (F) HS-65-96h aged at 100 C.
study, the successful preparation of these hierarchical materials mainly depended on the following parameters: - The Polymer:TEOS ratios (P:TEOS) in order to reach a sufficient volume fraction of polymer beads in the components of the solution - The aging conditions (time and temperature) to enable densification/aggregation of the organic-inorganic network.
3.2.1. Influence of the P:TEOS ratio on the mesoporous framework of SBA-15 A first series of hierarchical porous silica materials were prepared using different Polymer: TEOS ratios (P:TEOS). These are denoted HS-X-1.5, HS-X-2.5, HS-X-3.5 and HS-X-4.5 corresponding
respectively to 1.5, 2.5, 3.5 and 4.5: 1 weight ratios of polymer beads to TEOS. A 24 h hydrothermal treatment at 80 C was applied for all solids during the synthesis. Pure mesoporous SBA-15 was also prepared as reference for comparison using the same experimental conditions. The physical properties of these materials were assessed by SEM, N2 porosimetry and low-angle XRD, to observe the incorporation and structural periodicity of the macroporous and mesoporous networks. Macrostructure was first investigated by SEM. As illustrated Fig. 4A, SEM micrograph reveals that the undoped SBA-15 sample consists of a rod-like morphology characteristic of this class of material in such synthesis conditions. For all doped SBA-15 with a small amount of polymer beads (see Fig. 4C for HS-65-1.5 material), we can easily observe two distinct macrostructures corresponding to isolated SBA-15 rods and macroporous agglomerates coming from polymer spheres imprinting.
Table 2 Physical and textural properties of macroporous mesoporous silicas. Sample
S. Area/m2 g1
Total pore volumea/cm3 g1
Mesopore volumeb/cm3 g1
BJH pore diameterc/nm
Macropore sized/nm
d100e/nm
a/nm
Wall thicknessf/nm
SBA-15 HS-65-1.5 HS-65-3.5 HS-65-4.5 HS-100-1.5 HS-100-3.5 HS-200-1.5 HS-200-2.5 HS-200-3.5
800 739 743 642 651 705 578 525 607
0.83 1.16 1.78 1.35 1.25 1.50 0.66 0.67 0.74
0.74 1.09 1.71 1.34 1.21 1.43 0.59 0.62 0.67
5.2 4.5 4.3 4.3 4.2 4.3 5.3 4.0 4.3
e 58 46 e e 84 185 186 178
9.5 9 8.5 /(no diff) 8.0 7.9 8.3 8.4 7.9
10.9 10.4 9.9 / 9.2 9.1 9.6 9.7 9.1
5.7 5.9 5.5 / 5.0 4.8 4.8 4.7 4.8
a b c d e f
Total pore volume obtained at P/P0 ¼ 0.98. Mesopore volume from desorption isotherm using BJH model. Mesopore diameter from desorption isotherm using BJH model. Mean macropore diameter were estimated by a statistical analysis from TEM micrographs. Interplanar spacing determined from Bragg's law. Mesopore wall thickness calculated for a hexagonal system (2d100/√3)-Dmesopore.
146
A. Zaki et al. / Microporous and Mesoporous Materials 208 (2015) 140e151
Subsequent increase of the amount of macrotemplate progressively modifies the rod-like SBA-15 morphology and evolve into a more uniform macroporous network (Fig. 4B and D). The average diameter size of the cavities remains close to the parent bead dimension (10% shrinkage), suggesting that inorganic walls contraction has been attenuated by the slow ramp rate (0.5 C min1) used during the thermal treatment. However, despite some progress in the macrostructural organization obtained with optimised P:TEOS ratio (P:TEOS ¼ 3.5), it has been found that residual rods from the parent SBA-15 are still easily observed for HS-200-3.5 and particularly HS65-3.5 (Fig. 4B and D). This would originate from the incomplete sedimentation of such small polymer spheres. Classical sedimentation of small spherical objects (below 300 nm), depending on their density and on the viscosity of the solution, is a very long process [35,36] so that most of the works dealing within this size range have been done with pre-arranged colloidal dispersions of nanospheres before silicon alkoxide immersion/impregnation. Further increase of polymer sphere amount in the solution (P:TEOS ratio to 4.5) do not improve macropore incorporation but leads to the partial collapse of the macroporous scaffold. Then, assessment of mesoporosity for all materials has been performed by N2 porosimetry analysis (see Figs. 5 and 6) and relative textural informations have been summarized Table 2. As expected, the pure mesoporous SBA-15 exhibits a small steep adsorption increase at low P/P0 related to the microporosity coming from the residual presence of PEO chains in the silica walls. This result is well in line with the work of Choi et al. [37], where a low TEOS:P123 M ratio (z45) implied connections between the main channels of the mesoporous SBA-15. Then, capillary condensation phenomenon is occurring during the filling of pores at 0.40 < P/P0 < 0.60. A H1 hysteresis loop is observed with parallel adsorption and desorption branches that is typical of an ordered porous materials which exhibit a regular array of cylindrical pores. A specific surface area of 800 m2 g1 and a total pore volume of 0.83 cm3 g1 (Fig. 5A-a) have been obtained and are in line with literature [38]. For all doped silica materials, a combination of types IV and type II isotherms, indicating the presence of mesoporous and macroporous domains, have been observed with respect to SBA-15 following the macrotemplate size and amount used in the synthesis. H1 hysteresis loop size at 0.40 < P/P0 < 0.60 assigned to the well-organized pore structure of SBA-15 progressively decreases with increasing macropore character whatever the bead size, which may be accounted for by disruption of extended mesopore networks as previously reported [39,28]. In the high relative pressure domain (P/P0 ¼ 0.8), when large sphere sizes are used, N2 adsorption isotherm is only slightly perturbed and is revealing the intrinsic macroporosity of the HS-200 materials (Fig. 5A). As expected, multimolecular layers adsorption is increasing with the macroporous character of the material but contribute to decrease the surface area with respect to the mesoporous reference (see Table 2). Additional features have been found at high P/P0 range for HS-100 and HS-65 series (Fig. 6A), with the progressive formation of a hysteresis loop while increasing the P:TEOS ratio. This can be assigned to opened large mesopores originated from the interpenetrating macrotemplate walls. Thus, an additional independent mesoporous phase can be observed when using 100 nm beads. As a consequence, these large mesowindows allow to compensate the loss of mesoporous domains derived from the parent SBA-15 and to maintain a comparable surface area with respect to the pure mesoporous silica. Besides, such large mesopores increase significantly the pore volume of the material (1.78 cm3 g1 for HS-65-3.5 against 0.83 cm3 g1 for SBA-15). Finally, the structural periodicity of HS-solids with optimized P:TEOS ratio has been assessed by low angle XRD (Fig. 7). While the pure mesoporous SBA-15 material exhibits well defined peaks at 0.93 , 1.54 and 1.87, associated with the (100), (110) and (200)
Fig. 9. High magnification TEM micrographs of (A) HS-200-96h; (B) HS-100-96h and (C) HS-65-96h.
A. Zaki et al. / Microporous and Mesoporous Materials 208 (2015) 140e151
planes of the P6mm space group assigned to the hexagonal arrangement of mesopores, only one major reflection at 2q ¼ ~1 is obtained for all derived samples. Indubitably, macropore incorporation induces a partial loss of the long range ordered mesopore structure, in accordance with the broadening of the H1 hysteresis loop observed by N2 porosimetry. In addition, diffraction peaks are slightly shifting to higher angles, corresponding to a contraction in the mesopore lattice parameter whatever the bead size (see Table 2). Despite the parallel decrease of the lattice parameter and the mesopore size due to the presence of the macroporous
147
skeleton, hierarchical solids still maintain equivalent mesopore wall thickness with respect to the parent SBA-15 (Table 2). 3.2.2. Impact of aging conditions on SBA-15 macrostructure From above considerations, only one day aging under soft hydrothermal conditions is not sufficient to obtain a homogeneous arrangement of the composite material, thus generating a random packing or isolated macropores along with the presence of SBA-15 rod-like aggregates. In the following, the effect of aging conditions on the structure and textural properties will be discussed. A new
Fig. 10. N2 adsorption desorption isotherms of the hierarchical porous solids. (A) HS-200 series (B) HS-100 series,(C) HS-70 series and pore size distribution of (D) HS-200 series (E) HS-100 series,(F) HS-70 series with a ¼ 24 h, b ¼ 48 h and c ¼ 96 h.
148
A. Zaki et al. / Microporous and Mesoporous Materials 208 (2015) 140e151
series of macro-mesostructured materials with different aging time and temperature was prepared with a P:TEOS ¼ 3.5 which gave the best macrostructure organization. Prehydrolysed Si precursor/ P123/Polymer mixtures were maintained in a sealed polypropylene bottle under autogeneous pressure for a period of time ranging from 24 h to 96 h under static conditions at 80 C or 100 C. As illustrated Fig. 8, SEM micrographs reveal successful macropore incorporation for all siliceous materials whatever the aging time but only with a temperature of 80 C during the synthesis. For instance, it was clearly evidenced for HS-65 materials from SEM (Fig. 8D) and N2 porosimetry analysis (result not shown) that a temperature of 100 C contributes to the partial shrinkage of the macroporous skeleton. This latter phenomenon arose from the low glass transition temperature of PMMA polymer beads estimated from the DSC analysis, so that the hydrothermal treatment should be kept at temperature slightly lower than 100 C to prevent PMMA spheres from deformation. Besides, a well-ordered macrostructure is observed by expanding the aging time to 96h with a temperature fixed at 80 C (Fig. 8). It shows that a 96h hydrothermal treatment is clearly improving the close-packing of 200 nm and sub-100 nm nanospheres during which the solution penetrated into the interstices of polymer spheres through capillary force. TEM micrographs reveal the structural organization at both macroscale and mesoscale for siliceous materials (Fig. 9). As illustrated, when increasing aging time to 96h, interconnected macropores and mesopore networks respectively generated by the hard and soft templates are directly visualized on the samples. Such interconnected macropores are revealing the existence of mesoporous windows that connect each macropore, particularly when a long hydrothermal treatment is applied. The same phenomenon has been earlier observed by Stein et al. [5,40]. They found that the diameter of the windows can be tuned by sintering the spheres in the colloidal crystal together, forming necks that increase in diameter as the length of the thermal treatment is increased. In order to evaluate the textural changes implied by the mesowindows, the obtained materials have been characterized by N2 porosimetry. The sorption isotherms and corresponding pore-size distribution curves of HS solids aged at 80 C are shown Fig. 10. First, two condensation steps (0.40 < P/P0 < 0.70 and 0.70 < P/ P0 < 0.98) are evidenced for all HS solids which maintain a type IV shape albeit changes involved by aging time prolongation. Those features correspond respectively to the filling of small mesoporous channels with a pore size centered around 4 nm (see Table 3) generated from the P123 template and the filling of large mesoporous windows generated from the interpenetrated macropore
walls. While meso-windows only slightly perturbs the hysteresis at 0.40 < P/P0 < 0.70 for HS-200 solid (see Fig. 10A), reverse trends are clearly evidenced on the sub-100 nm macroporous mesoporous samples where the mesopore window contribution is dramatically increased (see Fig. 10). A parallel increase of the total pore volume is consequently observed on HS-100 and HS-65 materials. Thus, the pore volume and specific surface area of HS-100 and HS-65 samples calculated respectively from the condensation step and BET linear domain are significantly improved (see Table 3). HS-100-96 material, possessing 3 pore system networks, confers the highest surface area (983 m2 g1) and pore volume (2.66 cm3 g1) with respect to the pure mesoporous reference SBA-15 (800 m2 g1 and 0.83 cm3 g1). In addition, while HS-65-96 possess the smallest macropores (z50 nm), this material do not exhibit better textural properties than HS-100-96 as illustrated by the total pore volume obtained from the pore distribution curves (Table 3). This result could be related to SEM micrograph for HS-65-96 (see Fig. 8C) where amorphous regions without small macropores organization are observed. Finally, control on aging conditions over our hierarchical silicas materials is significantly modifying the mesopore wall thickness. Thicker walls are observed for all hierarchical silicas upon aging time (4.8 nm for HS-100-24 to 7.1 nm for HS-100-96 (5.7 nm for SBA-15)). In our case, this phenomenon is related to the pore size decrease of the material (from 5.2 nm for SBA-15 to z4 nm for HS materials) with the parallel stabilization of the lattice parameter with respect to the initial SBA-15. Hence, thicker pore wall would confer interesting hydrothermal properties to these hierarchical materials. Finally, macropores structural evolution as a function of aging time has been followed by means of SAXS. Fig. 11 depicts the integrated intensity profiles obtained for HS-200, HS100 and HS-65 samples aged during different times. For the HS-200 sample, no slopes changes on the scattered intensity are observed whatever the aging time. This arises from the fact that, due to their elevated size (around 200 nm), mesopores scattering is located in the very low q region that is not accessible by the equipment. For the HS-100-24 sample, a break of the slope is observed around q ¼ 0.1 nm-1. The corresponding size of the scattering object has been extracted from the data using the Unified fit approach developed by Ilavsky et al. [25], that gives access to the gyration radius of the scattering particle. Assuming that the scattering particles have a spherical shape, a characteristic diameter around 80 nm has been obtained, in agreement with TEM results (78 nm). Another indication given by the SAXS experiments is that the macropores are correlated or in other words that they are not randomly dispersed but rather interact the ones with the others.
Table 3 Physical and textural properties of optimized macroporous mesoporous silicas. Sample
Surface area/m2 g1
Total pore volumea/cm3 g1
Mesopore volumeb/cm3 g1
BJH pore diameterc/nm
Macropore sized/nm
d 100e/nm
A/nm
Wall thicknessf/nm
SBA-15 HS-200-24h HS-200-48h HS-200-96h HS-100-24h HS-100-48h HS-100-96h HS-65-24h HS-65-48h HS-65-96h
800 607 552 566 705 833 983 743 832 796
0.83 0.74 0.70 0.73 1.50 1.64 2.72 1.78 1.76 2.10
0.74 0.67 0.66 0.69 1.43 1.61 2.66 1.71 1.70 2.04
5.2 4.3 3.9 4.0 4.3 4.2 3.9 4.3 4.3 4.0
e 186 e 178 84 e 78 46 e 54
9.5 7.9 8.5 9.3 7.9 9.0 9.6 8.5 8.5 8.4
10.9 9.1 9.8 10.7 9.1 10.4 11.0 9.9 9.8 9.7
5.7 4.8 5.9 6.7 4.8 6.2 7.1 5.5 5.5 5.7
a b c d e f
(0.09) (0.07) (0.04) (0.04) (0.07) (0.03) (0.06) (0.07) (0.06) (0.06)
Total pore volume obtained at P/P0 ¼ 0.98. Mesopore volume from desorption isotherm using BJH model. Mesopore diameter from desorption isotherm using BJH model. Mean macropore diameter were estimated by a statistical analysis from TEM micrographs. Interplanar spacing determined from Bragg's law. Mesopore wall thickness calculated for a hexagonal system (2d100/√3)-Dmesopore.
A. Zaki et al. / Microporous and Mesoporous Materials 208 (2015) 140e151
Fig. 11. SAXS measurements at low q range of (A) HS-200 (B) HS-100 and (C) HS-65 samples at different hydrothermal aging time (a) 24 h (b) 48 h (c) 96 h.
149
150
A. Zaki et al. / Microporous and Mesoporous Materials 208 (2015) 140e151
However, upon aging, the scattering signal evolves and intensity profiles clearly indicate a loss of correlation between macropores. The same analyses have been performed on the HS-65 samples. In this case, a characteristic diameter around 50 nm has been obtained and is in adequation with electronic microscopy measurements (46 nm). Moreover for HS-65-24 sample a well-defined broad peak is observed on the scattered intensity at q ¼ 0.16 nm-1. This indicates that small macropores are strongly correlated and form a regular lattice. Interestingly, in the same way that for HS-100 samples, the correlation peak progressively vanish upon aging, showing again a loss of the correlation between macropores. These last results obtained on samples with sub-100 nm macropores are not in line with our previous electronic microscopy (SEM-TEM) observations where a better macrostructural ordering is clearly observed upon aging. This loss of correlation between macropores would arise in two complementary ways. First, while increasing the duration of the hydrothermal treatment, the large increase of mesoporous windows (ranging from 20 to 40 nm) induce a supplementary scattering signal very close to the small macropores (50 nm and 80 nm) in the low q region. This additional contribution would decrease the intensity profiles of the scattering signal and thus attenuating the break slope due to the macropores. In the other hand, the interconnectivity created between two adjacent macropores by large windows would imply the partial collapse of sub-100 nm macropores during the calcination step. This last explanation seems consistent with the textural results obtained on the hierarchical material presenting the smallest macropores (HS65 samples). While increasing the aging time from 24 h to 96 h, the specific surface area and pore volume of HS-65 samples are only slightly increased, indicating the limitation of aging treatment benefits for such small bead size. Finally, mesostructure periodicity of all materials has been assessed at high q range by SAXS. Indubitably, a broadening of the scattering peak at 0.7 nm1, corresponding to the small mesopore domain, is occurring while increasing hydrothermal aging time whatever the nanosphere template size. Close-packed macropores impact on the long range ordered mesopore structure, in accordance with N2 porosimetry. The decrease of mesopore ordering in our hierarchical silica materials can be related to the morphology changes of the P123 surfactant induced by the increase of the structural confinement using sub-100 nm polymer spheres with respect to large spheres [41]. In addition, the parallel intensification of amorphous regions around mesoporous windows, which originate from the exclusion of micelles formation near the polymer spheres, is particularly observed on the sub-100 nm hierarchical materials. All these results confirm the genesis of a partially ordered mesostructured phase for all materials (Fig. 12).
4. Conclusions In conclusion, we report a simple and straightforward method adapted from a sedimentation/aggregation technique to prepare hierarchical porous silica powders derived from SBA-15 with high surface area, high pore volume and three independent pore systems generated from polymer spheres with adjustable size, windows replicating from necks of polymer spheres arrays and a nonionic surfactant. We demonstrated that we can easily increase mesowindow entrances and mesopore wall thickness by simply adjusting the aging time using a hydrothermal treatment. The mesopore hexagonal network is obviously distorted by geometric constraints due to the macrostructured skeleton and smaller the bead is, smaller the mesopore domain remain within the interstitial voids of the sacrified nanospheres arrays. We expect that the new findings described in this paper will facilitate the
Fig. 12. SAXS measurements at high q range of (A) HS-200 (B) HS-100 and (C) HS-65 samples at different hydrothermal aging time (a) 24 h (b) 48 h (c) 96 h.
A. Zaki et al. / Microporous and Mesoporous Materials 208 (2015) 140e151
synthesis of tunable macroporous mesoporous derived from SBA15 materials. Acknowledgments gion The authors gratefully acknowledge the Contrat Plan Etat Re gion Nord Pas-de-Calais/FEDER for financial support. (CPER), the Re en de De veloppement Re gional (FEDER)”, The “Fonds Europe gion Nord Pas-de-Calais” and “Ministe re de l'Education “CNRS00 , “Re rieur et de la Recherche” are Nationale de l'Enseignement Supe acknowledged for fundings of X-ray diffractometers. References [1] C.M.A. Parlett, K. Wilson, A.F. Lee, Chem. Soc. Rev. 42 (2013) 3876. [2] T. Sen, G.J.T. Tiddy, J.L. Casci, M.W. Anderson, Angew. Chem. Int. Ed. 42 (2003) 4649. [3] D. Zhao, J. Feng, Q. Huo, N. Melosh, G.H. Fredrickson, B.F. Chmelka, G.D. Stucky, Science 279 (1998) 548. [4] Y. Wan, D.Y. Zhao, Chem. Rev. 107 (2007) 2821. [5] B.T. Holland, C.F. Blanford, T. Do, A. Stein, Chem. Mater. 11 (1999) 795. [6] X.Y. Yang, Y. Li, A. Lemaire, J.G. Yu, B.L. Su, Pure Appl. Chem. 81 (2009) 2265. [7] X. Li, Z.Y. Fu, B.L. Su, Adv. Funct. Mater. 22 (2012) 4634. [8] J. Sun, Z. Shan, T. Maschmeyer, J.A. Moulijn, M.O. Coppens, Chem. Comm. 24 (2001) 2670. [9] K. Na, C. Jo, J. Kim, K. Cho, J. Jung, Y. Seo, R.J. Messinger, B.F. Chmelka, R. Ryoo, Science 333 (2011) 328. rez-Ramírez, C.H. Christensen, K. Egeblad, C.H. Christensen, J.C. Groen, [10] J. Pe Chem. Soc. Rev. 37 (2008) 2530. [11] F. Schuth, Chem. Mater. 13 (2001) 3184. [12] B.T. Holland, C.F. Blanford, A. Stein, Science 281 (1998) 538. [13] S.A. Davis, S.L. Burkett, N.H. Mendelson, S. Mann, Nature 385 (1997) 420. [14] K.J. Nakanishi, SoleGel Sci. Tech. 19 (2000) 65.
151
n, K.K. Unger, [15] A. Lind, C.du Fresne von Hohenesche, J.H. Smatt, M. Linde Microporous Mesoporous Mater. 66 (2003) 219. n, Chem. Mater. 15 (2003) 2354. [16] J.H. Smatt, S. Schunk, M. Linde [17] T. Amatani, K. Nakanishi, K. Hirao, T. Kodaira, Chem. Mater. 17 (2005) 2114. [18] Z.G. Shia, Y.Q. Fenga, L. Xua, S.L. Daa, Y.-Y. Ren, Microporous Mesoporous Mater. 68 (2004) 55. [19] T. Arne, F. Goettmann, M. Antonietti, Chem. Mater. 20 (2008) 738. [20] O.D. Velev, T.A. Jede, R.F. Lobo, A.M. Lenhoff, Nature 389 (1997) 447. [21] Z. Whang, A. Stein, Chem. Mater. 20 (2008) 1029. [22] J. Zhao, P. Wan, J. Xiang, T. Tong, L. Dong, Z. Gao, X. Shen, H. Tong, Micro. meso. Mater. 138 (2011) 200. [23] A. Vu, A. Stein, Chem. Mater. 23 (2011) 3237. [24] S. Vaudreuil, M. Bousmina, S. Kaliaguine, L. Bonneviot, Adv. Mater. 13 (2001) 1310. [25] C.G. Oh, Y.K. Baek, S.K. Ihm, Adv. Mater. 7 (2005) 270. [26] J.S. Yun, S.K. Ihm, J. Phys. Chem. Solids 69 (2008) 1133. [27] J.J. Chiu, D.J. Pine, S.T. Bishop, B.F. Chmelka, J. Catal. 221 (2004) 400. [28] J. Dhainaut, J.P. Dacquin, A.F. Lee, K. Wilson, Green Chem. 12 (2010) 296. [29] C.M.A. Parlett, P. Keshwalla, S.G. Wainwright, D.W. Bruce, N.S. Hondow, K. Wilson, A.F. Lee, ACS Catal. 3 (2013) 2122. [30] Tongwen Wang, Ozlem Sel, Igor Djerdj, Bernd Smarsly, Colloid Polym. Sci. 285 (2006) 1. € ltner, H.-P. Hentze, Adv. Mater. 10 (1998) 154. [31] M. Antonietti, B. Berton, C. Go [32] J. Ilavsky, P.R. Jemian Irena, J. Appl. Cryst. 42 (2009) 347. [33] S.T. Camli, F. Buyukserin, M.S. Yavuz, G.G. Budak, J. Coll. Int. Sci. 344 (2010) 528. des Sciences et de Ge nie, Univ. Laval, Que bec, [34] S. Vaudreuil, PhD Thesis, Faculte 2003. [35] P. Pieranski, Contemp. Phys. 24 (1983) 25. [36] J.E.G. Wijnhoven, W.L. Vos, Science 281 (1998) 802. [37] M. Choi, W. Heo, F. Kleitz and R. Ryoo, 1340 (2003). [38] M. Kruk, M. Jaroniec, C. Hyun Ko, R. Ryoo, Chem. Mater. 12 (2000) 1961. [39] I. Nowak, Col. Surf. A 241 (2004) 103. [40] N.D. Petkovich, A. Stein, in: B.-L. Su, C. Sanchez, X.-Y. Yang (Eds.), Hierarchically Structured Porous Materials: from Nanoscience to Catalysis, Separation, Optics, Energy, and Life Science, vol. 1, Wiley-VCH Verlag GmbH and Co. KGaA, Weinheim, Germany, 2012, p. 55. [41] F. Li, Z. Wang, N.S. Ergang, C.A. Fyfe, A. Stein, Langmuir 23 (2007) 3996.
Contents lists available at ScienceDirect
Microporous and Mesoporous Materials journal homepage: www.elsevier.com/locate/micromeso
Tunable hierarchical porous silica materials using hydrothermal sedimentation-aggregation technique Abdelali Zaki a, Jing Xu a, Gregory Stoclet b, Sandra Casale c, Jean-Philippe Dacquin a, *, Pascal Granger a Unit e de Catalyse et Chimie du Solide (UCCS), CNRS UMR 8181, Universit e Lille I, Sciences et Technologies, 59655 Villeneuve d'Ascq, France Unit e Mat eriaux et Transformations (UMET), CNRS UMR 8207, Universit e Lille I, Sciences et Technologies, 59655 Villeneuve d'Ascq, France c Laboratoire de R eactivit e de Surface-CNRS UMR 7197, Universit e Pierre et Marie Curie, France a
b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 6 June 2014 Received in revised form 21 October 2014 Accepted 26 January 2015 Available online 3 February 2015
Herein, a simple and efficient synthesis based on a dual templating approach allow the preparation of tunable macroporous mesoporous silica materials. Macropores incorporation has been obtained using polymer spheres of well-defined size and their subsequent removal by thermal treatment allowed the independent control of the macropore size entrance between 200 nm and 50 nm. The addition of the block copolymer drives the formation of a second mesostructured skeleton, with a pore size centered around 4 nm, throughout the material framework. By a fine tuning of the solegel synthesis parameters (Polymer:TEOS ratio, aging conditions), we succeeded in guiding the SBA-15 rod-like morphology having randomly packed macropores to a derived SBA-15 with a homogeneous honeycomb macrostructure. Moreover, large mesoporous windows are generated between adjacent macropores. Hence, this simple one-pot synthesis approach, allowing scaling-up, offers fine tuning porosity at the macropore scale. © 2015 Elsevier Inc. All rights reserved.
Keywords: Hierarchical porous silicas SBA-15 Polymer spheres Hydrothermal sedimentaion-aggregation technique
1. Introduction Porous silica materials are widely used as supports for heterogeneous catalysis and separation applications since they offer various pore morphologies and high specific surface areas [1e5]. Moreover, with the emergence of alternative sources of energy, fuels and raw materials, the need for hierarchical architectures to prevent slow in-pore diffusion and poor atom efficiency of the reaction is highly desirable [6,7]. As a consequence, extensive researches on the design of heterogeneous solid supports have been developed since the end of the nineties [8e10] for maximizing the catalytic activity and reaching more sustainable technologies. In this line, templating strategies i.e. cooperative self-assembly [11], hard templates [12], biotemplates [13], phase separation [14] provide various possibilities for the synthesis of novel ordered mesostructures of tailored porosity and morphology. Various hierarchically structured silica materials can be obtained from the appropriate combination of such templating agents, including spray dried powders [15] and monoliths [16e18]. In these last
* Corresponding author. Tel.: þ33 (0)320434795. E-mail address: [email protected] (J.-P. Dacquin). http://dx.doi.org/10.1016/j.micromeso.2015.01.041 1387-1811/© 2015 Elsevier Inc. All rights reserved.
cases, surfactants such as cetylammonium salts or block copolymers can be added to induce a trimodal pore structure with micrometer-sized macropores. Dual-templating approach is as well an established technique and offers well-defined structural macroporosity and mesoporosity. Most often polymer spheres [19] exhibiting diameters of several hundred nanometers are used, and successful preparation of the macrostructured skeleton is crucially depending on the synthetic procedure usually based on the hard template pre-arrangement before infiltration by the surfactant-containing precursor [12,20]. This approach leads to the synthesis of hierarchical silica materials with an interconnected macropore network and mesopore network known as threedimensionally ordered macroporous materials with mesoporous walls (3 DOM/m materials) [21]. Research groups have also developed similar routes using polystyrene (PS) or polymethylmethacrylate (PMMA) spheres to produce 3 DOM/m transition metal oxides [22,23]. However, while the generation of smart and highly organized 3DOM structures can be obtained following this synthetic procedure, there is still restricted applications due to depth limitation in scaling-up [5]. Nevertheless, Ihm et al. proposed an alternative synthesis allowing feasible scaling-up with respect to the above described technique by using a solegel deposition of the silicate precursor on the polymer spheres
A. Zaki et al. / Microporous and Mesoporous Materials 208 (2015) 140e151
following hydrothermal sedimentation-aggregation technique [24] to generate a macroporous mesoporous MCM-41 [25]. This one-pot synthesis leads to a well-ordered macroporous skeleton with a mesoporous phase located between the walls of the macropores and having the same structure and arrangement of MCM-41. However, the direct application of this one pot synthesis to SBA15 materials, possessing better hydrothermal stability, led to a limited ordering of the macropores [26]. It has been found that the difference in the macrostructures between MCM-41 and SBA-15 originated from the difference in textural morphology of the parent mesoporous MCM-41 and SBA-15. Nevertheless, the incorporation of macropores (300 nm) in the mesoporous framework of the SBA-15 (5 nm) has been found to boost the catalytic activity for various relevant catalytic reactions [27e29]. Thus, the application of such macroporous mesoporous silica materials derived from SBA-15 demonstrated that an additional macroporous network could alleviate mass-transport limitations encountered in liquidphase reactions. Unfortunately, all those works [24e30] are reporting hierarchical porous silica materials with only one and large macropore size (300 nm). To our knowledge, hierarchical siliceous materials with sub<100 nm macropore size have been rarely reported in literature and are characterized by poorly ordered macrostructure with a broad macropore size distribution [31]. Herein, we report in a detailed study a simple and efficient one-pot synthesis allowing the preparation of hierarchical porous silica materials with sub-100 nm macropore size and bimodal mesoporosity. Our efforts were first devoted to the decrease of the polymer microsphere size (from 200 nm to 65 nm) and the research of right conditions to succeed in guiding the SBA-15 rod-like morphology having randomly packed macropores to a derived SBA15 having a well-organized honeycomb macrostructure with a mesoporous phase. Hence, improved textural properties have been found and could be highly desirable for liquid phase catalysis or adsorption of practical as well as fundamental interest. 2. Experimental section 2.1. Materials 2.1.1. Synthesis protocol of large polymer spheres (200 nm) Polystyrene spheres (PS) with a diameter size of 200 nm were prepared by an emulsifier-free emulsion polymerization technique and carried out in a 500 ml three neck bottom flask [24]. Typically, styrene (St) and divinylbenzene (DVB) are washed three times by an aqueous solution of NaOH (0.1 M) and then three times by distilled water in order to eliminate polymerization inhibitors. Then, 240 mL of distilled water is introduced in a 500 mL batch reactor. Before adding reagents, the aqueous solution is flushed under argon during at least 30 min in order to eliminate dissolved oxygen from the reactional medium. Then, 25.00 mL of St (monomer) and 4.75 mL of DVB (cross linker agent) are introduced in the reactor that is placed in a silicon oil bath under magnetic stirring and heating up to a fixed temperature (between 60 C and 90 C). After temperature stabilization, potassium persulfate (KPS) pre-dissolved in 20 mL of distilled water was injected to the reactional medium to initiate polymerization. The aqueous suspension of PS spheres was recovered after reaction by centrifugation or filtration and washed out with distilled water and dried at 80 C for 24 h. 2.1.2. Synthesis protocol of small polymer spheres (65 nm < PMMA < 100 nm) Polymethylmethacrylate spheres (PMMA) of diameter sizes between 65 nm and 100 nm were prepared by the surfactantassisted emulsion anionic polymerization technique [30] and carried out in a 500 mL three neck bottom flask. Typically, 330 mL of
141
distilled water is used as solvent and poured in a 500 mL three neck round bottom flask. Sodium dodecyl sulfate (SDS) surfactant and methyl methacrylate (MMA) monomers are added to the aqueous solution following 30 min of argon bubbling. Vigorous stirring is generally applied in order to have a good dispersion of MMA droplets and SDS. Then, the reactional medium is heated in a silicon oil bath at a desired temperature. After temperature stabilization, KPS pre-dissolved in 20 mL of distilled water was injected to the reactional medium to initiate polymerization. Argon bubbling is maintained during the experiment in order to limit undesired reactions which inhibit the polymer nodules growth. After 2 h reaction, the aqueous suspension of PMMA spheres is recovered by centrifugation and washed out with distilled water and dried at 80 C for 24 h. 2.1.3. Synthesis protocol of hierarchical porous silica materials (HS) Siliceous materials were synthesized following a similar dualtemplating route reported by Ihm et al. [26] Typically, 3.0 g of Pluronic P123 triblock copolymer was dissolved in 22.50 mL of water and 75.00 mL of 2 M HCl solution while stirring at 35 C. The required amount of polymer beads were then added to the solution and stirred for 1 h. Once a good dispersion of polymer beads was achieved, 6.90 mL of TEOS was added to the solution at 35 C and maintained for 24 h while stirring. The mixture was then transferred to the oven (T ¼ 80 C or 100 C) and maintained in a sealed polypropylene bottle under autogeneous pressure for a period of time ranging from 24 h to 96 h under static conditions. The solid product obtained is filtered, washed 3 times with deionised water and calcined in air at 550 C (ramp rate of 0.5 C min1) for 6 h. The final solids were labeled HS-X, where X ¼ polymer bead size. 2.1.4. Synthesis protocol of SBA-15 Pure mesoporous SBA-15 silica was synthesized following the approach of Zhao et al. [3] Typically, 3.0 g of Pluronic P123 triblock copolymer was dissolved in 22.50 mL of water and 75 mL of 2 M HCl solution while stirring at 35 C. 6.90 mL of TEOS (TEOS:P123 M ratio of 40) were added to the solution, which was maintained at 35 C for 24 h while stirring. The mixture was then aged at 80 C for 24 h and the solid product filtered, washed 3 times with deionised water, and calcined in air at 550 C (ramp rate of 0.5 C min1) for 6 h. 2.2. Characterisation techniques The average hydrodynamic diameters of PS and PMMA spheres were determined by dynamic light scattering (DLS, Nano ZS, Malvern Instruments). DLS was applied with an angle of 173 by using HeeNe laser (4 mW) operated at 633 nm. Average hydrodynamic diameters of nanoparticles were measured in aqueous solutions. Thermo-gravimetric analysis (TGA) was performed to study the thermal decomposition of polymer spheres. A SDT2960 TA instrument was used. 20 mg of sample were placed in an alumina crucible with a volume of 100 ml. The thermal analyzes were performed under air. The sample heated with a ramp in temperature of 5 C min1, from 25 to 700 C. Textural properties were carried out on an automated gas adsorption analyzer (Tristar 3020). Nitrogen sorption isotherms of the calcined samples were measured after outgassing at 200 C in vacuum for 3 h. The multipoint surface area was evaluated with the BrunauereEmmetteTeller (BET) method over the range P/P0 ¼ 0.075e0.35 and pore size distribution was obtained using BarretteJoynereHalenda (BJH) model applied to the desorption isotherm branch. Total pore volume was determined from the volume adsorbed at P/P0 ¼ 0.98. Transmission Electron micrographs were collected on a JEOL JEM 2011 (LaB6) operating at 200 kV while scanning emission
142
A. Zaki et al. / Microporous and Mesoporous Materials 208 (2015) 140e151
Table 1 Size controlled synthesis of anionic polymer nanospheres Sample
Monomer/mL
KPS/g
SDS/g
Stirring/rpm
Time reaction/h
Temperature/ C
dhyd/nm
dp/nm
PS-200 PS-145 PMMA-100 PMMA-65
25 24 24 24
0.083 0.187 0.187 0.187
/ 0.476 0.117 0.238
200 450 450 450
5 5 2 2
80 80 80 80
225 150 130 82
200 145 100 65
microscopy (Hitachi SU-70, SEM-FEG) was used for microstructure analysis of siliceous materials. Low angle X-Ray Diffraction (XRD) measurements was performed using a D8 advanced X-Ray diffractometer (Bruker AXS) equipped with a Cu-Ka (l ¼ 0.154 nm) radiation. Low angle XRD patterns were acquired from 0.3 to 6.0 with a 0.02 steps at 10 s per point. SAXS experiments were carried out using a Genix microsource (Xenocs, France) operated at 50 kV and 1 mA. The Cu-Ka radiation (l ¼ 0.154 nm) was selected and focused by means of FOX2D mirrors. The sample to detector distance was varied between 0.5 and 2 m. Silver Behenate was used as calibrant for the precise determination of the sample to detector distance. The 2D-SAXS patterns were recorded on a Photonic Science VHR CCD camera. Corrections were applied for background scattering, geometry and intensity distortions of the detector. Integrated intensity profiles I ¼ f(q), where q is the scattering vector (q¼(4p/l sin(q)/2)), were computed using the Fit2D software. Fitting of the scattered intensities I ¼ f(q) was performed using the Irena Package [32].
Fig. 1. Scanning electron images of (a) PS-200 and (b) PM-65 polymer spheres.
± ± ± ±
24 9 10 7
Yield/% 57 35 74 84
3. Results and discussions 3.1. Macrotemplate characterization Prior to the synthesis of the siliceous architectures, polymer spheres have been prepared using the emulsion polymerization technique. Large PS spheres have been successfully obtained following the surfactant-free emulsion polymerization which is one of the most common methods used for the production of polymers. Besides, by adding acetone surfactant within the mixture, PS sphere size can easily fall down according to literature [33]. However, only poor yields have been obtained in this study using St as monomer (Table 1). In order to reach lower sphere size while keeping a significant synthesis yield, the SDS surfactant assisted emulsion technique has been used to produce small PMMA nanospheres [30]. Results gathered in Table 1 and Fig. 1 show that nearly monodispersed spherical polymer spheres have been prepared with yields >50% and controllable size between 200 nm and 65 nm without altering the morphological properties. Thermal behavior of the macrotemplates (PMMA and PS) has been examined by thermogravimetric analysis (TGA) under air. Fig. 2
Fig. 2. TGA thermograms of weight loss and derivative weight loss as a function of temperature. (a) PM-65 and (b) PS-200 polymer spheres.
A. Zaki et al. / Microporous and Mesoporous Materials 208 (2015) 140e151
143
Fig. 3. Schematic illustration of the preparation of hierarchical porous silica by dual templating route.
Fig. 4. SEM micrographs of SBA-15 reference (a) and selected hierarchical porous silica materials with different Polymer:TEOS ratios (b) HS-200-3.5, (c) HS-65-1.5 and (d) HS-653.5.
Fig. 5. N2 adsorption desorption isotherms (A) and pore size distribution (B) of SBA-15 and macroporous mesoporous solids with different P/TEOS ratio (a) SBA-15, (b) HS-200-1.5, (c) HS-200-2.5 and (d) HS-200-3.5.
144
A. Zaki et al. / Microporous and Mesoporous Materials 208 (2015) 140e151
shows the sample weight loss and respective derivative weight loss as a function of temperature for PS and PMMA spheres (Fig. 2). As earlier evidenced in literature, the onset of the thermal decomposition of PS beads (287 C) is shifted toward the higher temperature range than that of PMMA beads (261 C), which confirms the enhancement of thermal stability of PS template. However, while the decomposition of PMMA proceeds in one step with a complete combustion of the organic material achieved at 350 C, the decomposition of PS spheres involves a two oxidation step process [34]. As a consequence, total weight loss of PS spheres is only achieved at 515 C. Based on these observations, the calcination temperature of hierarchical porous materials was set at 550 C.
3.2. Synthesis of tunable macroporous mesoporous silica materials Our dual-templating method involves the direct insertion of the macrotemplate in the solution following the addition of the TEOS and block-copolymer (Fig. 3). With a pH below 2, the solution containing the prehydrolysed precursor subsequently impregnate the anionic polymer spheres during one day and silica condensation reaction is achieved following hydrothermal treatment, resulting in an intermediate composite nanospheres/P123-silica structure. Hybrid materials were calcined in order to remove the organic porogens and to obtain the final siliceous solids. During this
Fig. 7. Low angle X-ray diffraction of SBA-15 reference and HS solids with optimised P:TEOS ratio. (a) SBA-15, (b) HS-65-3.5, (c) HS-100-3.5, (d) HS-200-3.5.
Fig. 6. N2 adsorption desorption isotherms (A) and pore size distribution (B) of macroporous mesoporous solids with different P/TEOS ratio. (a) HS-65-1.5 (b) HS-65-3.5, (c) HS-654.5, (d) HS-100-1.5, (e) HS-100-3.5.
A. Zaki et al. / Microporous and Mesoporous Materials 208 (2015) 140e151
145
Fig. 8. SEM micrographs of porous samples (A) HS-200-96h aged at 80 C; (B) HS-100-96h aged at 80 C and (C) HS-65-96h aged at 80 C (F) HS-65-96h aged at 100 C.
study, the successful preparation of these hierarchical materials mainly depended on the following parameters: - The Polymer:TEOS ratios (P:TEOS) in order to reach a sufficient volume fraction of polymer beads in the components of the solution - The aging conditions (time and temperature) to enable densification/aggregation of the organic-inorganic network.
3.2.1. Influence of the P:TEOS ratio on the mesoporous framework of SBA-15 A first series of hierarchical porous silica materials were prepared using different Polymer: TEOS ratios (P:TEOS). These are denoted HS-X-1.5, HS-X-2.5, HS-X-3.5 and HS-X-4.5 corresponding
respectively to 1.5, 2.5, 3.5 and 4.5: 1 weight ratios of polymer beads to TEOS. A 24 h hydrothermal treatment at 80 C was applied for all solids during the synthesis. Pure mesoporous SBA-15 was also prepared as reference for comparison using the same experimental conditions. The physical properties of these materials were assessed by SEM, N2 porosimetry and low-angle XRD, to observe the incorporation and structural periodicity of the macroporous and mesoporous networks. Macrostructure was first investigated by SEM. As illustrated Fig. 4A, SEM micrograph reveals that the undoped SBA-15 sample consists of a rod-like morphology characteristic of this class of material in such synthesis conditions. For all doped SBA-15 with a small amount of polymer beads (see Fig. 4C for HS-65-1.5 material), we can easily observe two distinct macrostructures corresponding to isolated SBA-15 rods and macroporous agglomerates coming from polymer spheres imprinting.
Table 2 Physical and textural properties of macroporous mesoporous silicas. Sample
S. Area/m2 g1
Total pore volumea/cm3 g1
Mesopore volumeb/cm3 g1
BJH pore diameterc/nm
Macropore sized/nm
d100e/nm
a/nm
Wall thicknessf/nm
SBA-15 HS-65-1.5 HS-65-3.5 HS-65-4.5 HS-100-1.5 HS-100-3.5 HS-200-1.5 HS-200-2.5 HS-200-3.5
800 739 743 642 651 705 578 525 607
0.83 1.16 1.78 1.35 1.25 1.50 0.66 0.67 0.74
0.74 1.09 1.71 1.34 1.21 1.43 0.59 0.62 0.67
5.2 4.5 4.3 4.3 4.2 4.3 5.3 4.0 4.3
e 58 46 e e 84 185 186 178
9.5 9 8.5 /(no diff) 8.0 7.9 8.3 8.4 7.9
10.9 10.4 9.9 / 9.2 9.1 9.6 9.7 9.1
5.7 5.9 5.5 / 5.0 4.8 4.8 4.7 4.8
a b c d e f
Total pore volume obtained at P/P0 ¼ 0.98. Mesopore volume from desorption isotherm using BJH model. Mesopore diameter from desorption isotherm using BJH model. Mean macropore diameter were estimated by a statistical analysis from TEM micrographs. Interplanar spacing determined from Bragg's law. Mesopore wall thickness calculated for a hexagonal system (2d100/√3)-Dmesopore.
146
A. Zaki et al. / Microporous and Mesoporous Materials 208 (2015) 140e151
Subsequent increase of the amount of macrotemplate progressively modifies the rod-like SBA-15 morphology and evolve into a more uniform macroporous network (Fig. 4B and D). The average diameter size of the cavities remains close to the parent bead dimension (10% shrinkage), suggesting that inorganic walls contraction has been attenuated by the slow ramp rate (0.5 C min1) used during the thermal treatment. However, despite some progress in the macrostructural organization obtained with optimised P:TEOS ratio (P:TEOS ¼ 3.5), it has been found that residual rods from the parent SBA-15 are still easily observed for HS-200-3.5 and particularly HS65-3.5 (Fig. 4B and D). This would originate from the incomplete sedimentation of such small polymer spheres. Classical sedimentation of small spherical objects (below 300 nm), depending on their density and on the viscosity of the solution, is a very long process [35,36] so that most of the works dealing within this size range have been done with pre-arranged colloidal dispersions of nanospheres before silicon alkoxide immersion/impregnation. Further increase of polymer sphere amount in the solution (P:TEOS ratio to 4.5) do not improve macropore incorporation but leads to the partial collapse of the macroporous scaffold. Then, assessment of mesoporosity for all materials has been performed by N2 porosimetry analysis (see Figs. 5 and 6) and relative textural informations have been summarized Table 2. As expected, the pure mesoporous SBA-15 exhibits a small steep adsorption increase at low P/P0 related to the microporosity coming from the residual presence of PEO chains in the silica walls. This result is well in line with the work of Choi et al. [37], where a low TEOS:P123 M ratio (z45) implied connections between the main channels of the mesoporous SBA-15. Then, capillary condensation phenomenon is occurring during the filling of pores at 0.40 < P/P0 < 0.60. A H1 hysteresis loop is observed with parallel adsorption and desorption branches that is typical of an ordered porous materials which exhibit a regular array of cylindrical pores. A specific surface area of 800 m2 g1 and a total pore volume of 0.83 cm3 g1 (Fig. 5A-a) have been obtained and are in line with literature [38]. For all doped silica materials, a combination of types IV and type II isotherms, indicating the presence of mesoporous and macroporous domains, have been observed with respect to SBA-15 following the macrotemplate size and amount used in the synthesis. H1 hysteresis loop size at 0.40 < P/P0 < 0.60 assigned to the well-organized pore structure of SBA-15 progressively decreases with increasing macropore character whatever the bead size, which may be accounted for by disruption of extended mesopore networks as previously reported [39,28]. In the high relative pressure domain (P/P0 ¼ 0.8), when large sphere sizes are used, N2 adsorption isotherm is only slightly perturbed and is revealing the intrinsic macroporosity of the HS-200 materials (Fig. 5A). As expected, multimolecular layers adsorption is increasing with the macroporous character of the material but contribute to decrease the surface area with respect to the mesoporous reference (see Table 2). Additional features have been found at high P/P0 range for HS-100 and HS-65 series (Fig. 6A), with the progressive formation of a hysteresis loop while increasing the P:TEOS ratio. This can be assigned to opened large mesopores originated from the interpenetrating macrotemplate walls. Thus, an additional independent mesoporous phase can be observed when using 100 nm beads. As a consequence, these large mesowindows allow to compensate the loss of mesoporous domains derived from the parent SBA-15 and to maintain a comparable surface area with respect to the pure mesoporous silica. Besides, such large mesopores increase significantly the pore volume of the material (1.78 cm3 g1 for HS-65-3.5 against 0.83 cm3 g1 for SBA-15). Finally, the structural periodicity of HS-solids with optimized P:TEOS ratio has been assessed by low angle XRD (Fig. 7). While the pure mesoporous SBA-15 material exhibits well defined peaks at 0.93 , 1.54 and 1.87, associated with the (100), (110) and (200)
Fig. 9. High magnification TEM micrographs of (A) HS-200-96h; (B) HS-100-96h and (C) HS-65-96h.
A. Zaki et al. / Microporous and Mesoporous Materials 208 (2015) 140e151
planes of the P6mm space group assigned to the hexagonal arrangement of mesopores, only one major reflection at 2q ¼ ~1 is obtained for all derived samples. Indubitably, macropore incorporation induces a partial loss of the long range ordered mesopore structure, in accordance with the broadening of the H1 hysteresis loop observed by N2 porosimetry. In addition, diffraction peaks are slightly shifting to higher angles, corresponding to a contraction in the mesopore lattice parameter whatever the bead size (see Table 2). Despite the parallel decrease of the lattice parameter and the mesopore size due to the presence of the macroporous
147
skeleton, hierarchical solids still maintain equivalent mesopore wall thickness with respect to the parent SBA-15 (Table 2). 3.2.2. Impact of aging conditions on SBA-15 macrostructure From above considerations, only one day aging under soft hydrothermal conditions is not sufficient to obtain a homogeneous arrangement of the composite material, thus generating a random packing or isolated macropores along with the presence of SBA-15 rod-like aggregates. In the following, the effect of aging conditions on the structure and textural properties will be discussed. A new
Fig. 10. N2 adsorption desorption isotherms of the hierarchical porous solids. (A) HS-200 series (B) HS-100 series,(C) HS-70 series and pore size distribution of (D) HS-200 series (E) HS-100 series,(F) HS-70 series with a ¼ 24 h, b ¼ 48 h and c ¼ 96 h.
148
A. Zaki et al. / Microporous and Mesoporous Materials 208 (2015) 140e151
series of macro-mesostructured materials with different aging time and temperature was prepared with a P:TEOS ¼ 3.5 which gave the best macrostructure organization. Prehydrolysed Si precursor/ P123/Polymer mixtures were maintained in a sealed polypropylene bottle under autogeneous pressure for a period of time ranging from 24 h to 96 h under static conditions at 80 C or 100 C. As illustrated Fig. 8, SEM micrographs reveal successful macropore incorporation for all siliceous materials whatever the aging time but only with a temperature of 80 C during the synthesis. For instance, it was clearly evidenced for HS-65 materials from SEM (Fig. 8D) and N2 porosimetry analysis (result not shown) that a temperature of 100 C contributes to the partial shrinkage of the macroporous skeleton. This latter phenomenon arose from the low glass transition temperature of PMMA polymer beads estimated from the DSC analysis, so that the hydrothermal treatment should be kept at temperature slightly lower than 100 C to prevent PMMA spheres from deformation. Besides, a well-ordered macrostructure is observed by expanding the aging time to 96h with a temperature fixed at 80 C (Fig. 8). It shows that a 96h hydrothermal treatment is clearly improving the close-packing of 200 nm and sub-100 nm nanospheres during which the solution penetrated into the interstices of polymer spheres through capillary force. TEM micrographs reveal the structural organization at both macroscale and mesoscale for siliceous materials (Fig. 9). As illustrated, when increasing aging time to 96h, interconnected macropores and mesopore networks respectively generated by the hard and soft templates are directly visualized on the samples. Such interconnected macropores are revealing the existence of mesoporous windows that connect each macropore, particularly when a long hydrothermal treatment is applied. The same phenomenon has been earlier observed by Stein et al. [5,40]. They found that the diameter of the windows can be tuned by sintering the spheres in the colloidal crystal together, forming necks that increase in diameter as the length of the thermal treatment is increased. In order to evaluate the textural changes implied by the mesowindows, the obtained materials have been characterized by N2 porosimetry. The sorption isotherms and corresponding pore-size distribution curves of HS solids aged at 80 C are shown Fig. 10. First, two condensation steps (0.40 < P/P0 < 0.70 and 0.70 < P/ P0 < 0.98) are evidenced for all HS solids which maintain a type IV shape albeit changes involved by aging time prolongation. Those features correspond respectively to the filling of small mesoporous channels with a pore size centered around 4 nm (see Table 3) generated from the P123 template and the filling of large mesoporous windows generated from the interpenetrated macropore
walls. While meso-windows only slightly perturbs the hysteresis at 0.40 < P/P0 < 0.70 for HS-200 solid (see Fig. 10A), reverse trends are clearly evidenced on the sub-100 nm macroporous mesoporous samples where the mesopore window contribution is dramatically increased (see Fig. 10). A parallel increase of the total pore volume is consequently observed on HS-100 and HS-65 materials. Thus, the pore volume and specific surface area of HS-100 and HS-65 samples calculated respectively from the condensation step and BET linear domain are significantly improved (see Table 3). HS-100-96 material, possessing 3 pore system networks, confers the highest surface area (983 m2 g1) and pore volume (2.66 cm3 g1) with respect to the pure mesoporous reference SBA-15 (800 m2 g1 and 0.83 cm3 g1). In addition, while HS-65-96 possess the smallest macropores (z50 nm), this material do not exhibit better textural properties than HS-100-96 as illustrated by the total pore volume obtained from the pore distribution curves (Table 3). This result could be related to SEM micrograph for HS-65-96 (see Fig. 8C) where amorphous regions without small macropores organization are observed. Finally, control on aging conditions over our hierarchical silicas materials is significantly modifying the mesopore wall thickness. Thicker walls are observed for all hierarchical silicas upon aging time (4.8 nm for HS-100-24 to 7.1 nm for HS-100-96 (5.7 nm for SBA-15)). In our case, this phenomenon is related to the pore size decrease of the material (from 5.2 nm for SBA-15 to z4 nm for HS materials) with the parallel stabilization of the lattice parameter with respect to the initial SBA-15. Hence, thicker pore wall would confer interesting hydrothermal properties to these hierarchical materials. Finally, macropores structural evolution as a function of aging time has been followed by means of SAXS. Fig. 11 depicts the integrated intensity profiles obtained for HS-200, HS100 and HS-65 samples aged during different times. For the HS-200 sample, no slopes changes on the scattered intensity are observed whatever the aging time. This arises from the fact that, due to their elevated size (around 200 nm), mesopores scattering is located in the very low q region that is not accessible by the equipment. For the HS-100-24 sample, a break of the slope is observed around q ¼ 0.1 nm-1. The corresponding size of the scattering object has been extracted from the data using the Unified fit approach developed by Ilavsky et al. [25], that gives access to the gyration radius of the scattering particle. Assuming that the scattering particles have a spherical shape, a characteristic diameter around 80 nm has been obtained, in agreement with TEM results (78 nm). Another indication given by the SAXS experiments is that the macropores are correlated or in other words that they are not randomly dispersed but rather interact the ones with the others.
Table 3 Physical and textural properties of optimized macroporous mesoporous silicas. Sample
Surface area/m2 g1
Total pore volumea/cm3 g1
Mesopore volumeb/cm3 g1
BJH pore diameterc/nm
Macropore sized/nm
d 100e/nm
A/nm
Wall thicknessf/nm
SBA-15 HS-200-24h HS-200-48h HS-200-96h HS-100-24h HS-100-48h HS-100-96h HS-65-24h HS-65-48h HS-65-96h
800 607 552 566 705 833 983 743 832 796
0.83 0.74 0.70 0.73 1.50 1.64 2.72 1.78 1.76 2.10
0.74 0.67 0.66 0.69 1.43 1.61 2.66 1.71 1.70 2.04
5.2 4.3 3.9 4.0 4.3 4.2 3.9 4.3 4.3 4.0
e 186 e 178 84 e 78 46 e 54
9.5 7.9 8.5 9.3 7.9 9.0 9.6 8.5 8.5 8.4
10.9 9.1 9.8 10.7 9.1 10.4 11.0 9.9 9.8 9.7
5.7 4.8 5.9 6.7 4.8 6.2 7.1 5.5 5.5 5.7
a b c d e f
(0.09) (0.07) (0.04) (0.04) (0.07) (0.03) (0.06) (0.07) (0.06) (0.06)
Total pore volume obtained at P/P0 ¼ 0.98. Mesopore volume from desorption isotherm using BJH model. Mesopore diameter from desorption isotherm using BJH model. Mean macropore diameter were estimated by a statistical analysis from TEM micrographs. Interplanar spacing determined from Bragg's law. Mesopore wall thickness calculated for a hexagonal system (2d100/√3)-Dmesopore.
A. Zaki et al. / Microporous and Mesoporous Materials 208 (2015) 140e151
Fig. 11. SAXS measurements at low q range of (A) HS-200 (B) HS-100 and (C) HS-65 samples at different hydrothermal aging time (a) 24 h (b) 48 h (c) 96 h.
149
150
A. Zaki et al. / Microporous and Mesoporous Materials 208 (2015) 140e151
However, upon aging, the scattering signal evolves and intensity profiles clearly indicate a loss of correlation between macropores. The same analyses have been performed on the HS-65 samples. In this case, a characteristic diameter around 50 nm has been obtained and is in adequation with electronic microscopy measurements (46 nm). Moreover for HS-65-24 sample a well-defined broad peak is observed on the scattered intensity at q ¼ 0.16 nm-1. This indicates that small macropores are strongly correlated and form a regular lattice. Interestingly, in the same way that for HS-100 samples, the correlation peak progressively vanish upon aging, showing again a loss of the correlation between macropores. These last results obtained on samples with sub-100 nm macropores are not in line with our previous electronic microscopy (SEM-TEM) observations where a better macrostructural ordering is clearly observed upon aging. This loss of correlation between macropores would arise in two complementary ways. First, while increasing the duration of the hydrothermal treatment, the large increase of mesoporous windows (ranging from 20 to 40 nm) induce a supplementary scattering signal very close to the small macropores (50 nm and 80 nm) in the low q region. This additional contribution would decrease the intensity profiles of the scattering signal and thus attenuating the break slope due to the macropores. In the other hand, the interconnectivity created between two adjacent macropores by large windows would imply the partial collapse of sub-100 nm macropores during the calcination step. This last explanation seems consistent with the textural results obtained on the hierarchical material presenting the smallest macropores (HS65 samples). While increasing the aging time from 24 h to 96 h, the specific surface area and pore volume of HS-65 samples are only slightly increased, indicating the limitation of aging treatment benefits for such small bead size. Finally, mesostructure periodicity of all materials has been assessed at high q range by SAXS. Indubitably, a broadening of the scattering peak at 0.7 nm1, corresponding to the small mesopore domain, is occurring while increasing hydrothermal aging time whatever the nanosphere template size. Close-packed macropores impact on the long range ordered mesopore structure, in accordance with N2 porosimetry. The decrease of mesopore ordering in our hierarchical silica materials can be related to the morphology changes of the P123 surfactant induced by the increase of the structural confinement using sub-100 nm polymer spheres with respect to large spheres [41]. In addition, the parallel intensification of amorphous regions around mesoporous windows, which originate from the exclusion of micelles formation near the polymer spheres, is particularly observed on the sub-100 nm hierarchical materials. All these results confirm the genesis of a partially ordered mesostructured phase for all materials (Fig. 12).
4. Conclusions In conclusion, we report a simple and straightforward method adapted from a sedimentation/aggregation technique to prepare hierarchical porous silica powders derived from SBA-15 with high surface area, high pore volume and three independent pore systems generated from polymer spheres with adjustable size, windows replicating from necks of polymer spheres arrays and a nonionic surfactant. We demonstrated that we can easily increase mesowindow entrances and mesopore wall thickness by simply adjusting the aging time using a hydrothermal treatment. The mesopore hexagonal network is obviously distorted by geometric constraints due to the macrostructured skeleton and smaller the bead is, smaller the mesopore domain remain within the interstitial voids of the sacrified nanospheres arrays. We expect that the new findings described in this paper will facilitate the
Fig. 12. SAXS measurements at high q range of (A) HS-200 (B) HS-100 and (C) HS-65 samples at different hydrothermal aging time (a) 24 h (b) 48 h (c) 96 h.
A. Zaki et al. / Microporous and Mesoporous Materials 208 (2015) 140e151
synthesis of tunable macroporous mesoporous derived from SBA15 materials. Acknowledgments gion The authors gratefully acknowledge the Contrat Plan Etat Re gion Nord Pas-de-Calais/FEDER for financial support. (CPER), the Re en de De veloppement Re gional (FEDER)”, The “Fonds Europe gion Nord Pas-de-Calais” and “Ministe re de l'Education “CNRS00 , “Re rieur et de la Recherche” are Nationale de l'Enseignement Supe acknowledged for fundings of X-ray diffractometers. References [1] C.M.A. Parlett, K. Wilson, A.F. Lee, Chem. Soc. Rev. 42 (2013) 3876. [2] T. Sen, G.J.T. Tiddy, J.L. Casci, M.W. Anderson, Angew. Chem. Int. Ed. 42 (2003) 4649. [3] D. Zhao, J. Feng, Q. Huo, N. Melosh, G.H. Fredrickson, B.F. Chmelka, G.D. Stucky, Science 279 (1998) 548. [4] Y. Wan, D.Y. Zhao, Chem. Rev. 107 (2007) 2821. [5] B.T. Holland, C.F. Blanford, T. Do, A. Stein, Chem. Mater. 11 (1999) 795. [6] X.Y. Yang, Y. Li, A. Lemaire, J.G. Yu, B.L. Su, Pure Appl. Chem. 81 (2009) 2265. [7] X. Li, Z.Y. Fu, B.L. Su, Adv. Funct. Mater. 22 (2012) 4634. [8] J. Sun, Z. Shan, T. Maschmeyer, J.A. Moulijn, M.O. Coppens, Chem. Comm. 24 (2001) 2670. [9] K. Na, C. Jo, J. Kim, K. Cho, J. Jung, Y. Seo, R.J. Messinger, B.F. Chmelka, R. Ryoo, Science 333 (2011) 328. rez-Ramírez, C.H. Christensen, K. Egeblad, C.H. Christensen, J.C. Groen, [10] J. Pe Chem. Soc. Rev. 37 (2008) 2530. [11] F. Schuth, Chem. Mater. 13 (2001) 3184. [12] B.T. Holland, C.F. Blanford, A. Stein, Science 281 (1998) 538. [13] S.A. Davis, S.L. Burkett, N.H. Mendelson, S. Mann, Nature 385 (1997) 420. [14] K.J. Nakanishi, SoleGel Sci. Tech. 19 (2000) 65.
151
n, K.K. Unger, [15] A. Lind, C.du Fresne von Hohenesche, J.H. Smatt, M. Linde Microporous Mesoporous Mater. 66 (2003) 219. n, Chem. Mater. 15 (2003) 2354. [16] J.H. Smatt, S. Schunk, M. Linde [17] T. Amatani, K. Nakanishi, K. Hirao, T. Kodaira, Chem. Mater. 17 (2005) 2114. [18] Z.G. Shia, Y.Q. Fenga, L. Xua, S.L. Daa, Y.-Y. Ren, Microporous Mesoporous Mater. 68 (2004) 55. [19] T. Arne, F. Goettmann, M. Antonietti, Chem. Mater. 20 (2008) 738. [20] O.D. Velev, T.A. Jede, R.F. Lobo, A.M. Lenhoff, Nature 389 (1997) 447. [21] Z. Whang, A. Stein, Chem. Mater. 20 (2008) 1029. [22] J. Zhao, P. Wan, J. Xiang, T. Tong, L. Dong, Z. Gao, X. Shen, H. Tong, Micro. meso. Mater. 138 (2011) 200. [23] A. Vu, A. Stein, Chem. Mater. 23 (2011) 3237. [24] S. Vaudreuil, M. Bousmina, S. Kaliaguine, L. Bonneviot, Adv. Mater. 13 (2001) 1310. [25] C.G. Oh, Y.K. Baek, S.K. Ihm, Adv. Mater. 7 (2005) 270. [26] J.S. Yun, S.K. Ihm, J. Phys. Chem. Solids 69 (2008) 1133. [27] J.J. Chiu, D.J. Pine, S.T. Bishop, B.F. Chmelka, J. Catal. 221 (2004) 400. [28] J. Dhainaut, J.P. Dacquin, A.F. Lee, K. Wilson, Green Chem. 12 (2010) 296. [29] C.M.A. Parlett, P. Keshwalla, S.G. Wainwright, D.W. Bruce, N.S. Hondow, K. Wilson, A.F. Lee, ACS Catal. 3 (2013) 2122. [30] Tongwen Wang, Ozlem Sel, Igor Djerdj, Bernd Smarsly, Colloid Polym. Sci. 285 (2006) 1. € ltner, H.-P. Hentze, Adv. Mater. 10 (1998) 154. [31] M. Antonietti, B. Berton, C. Go [32] J. Ilavsky, P.R. Jemian Irena, J. Appl. Cryst. 42 (2009) 347. [33] S.T. Camli, F. Buyukserin, M.S. Yavuz, G.G. Budak, J. Coll. Int. Sci. 344 (2010) 528. des Sciences et de Ge nie, Univ. Laval, Que bec, [34] S. Vaudreuil, PhD Thesis, Faculte 2003. [35] P. Pieranski, Contemp. Phys. 24 (1983) 25. [36] J.E.G. Wijnhoven, W.L. Vos, Science 281 (1998) 802. [37] M. Choi, W. Heo, F. Kleitz and R. Ryoo, 1340 (2003). [38] M. Kruk, M. Jaroniec, C. Hyun Ko, R. Ryoo, Chem. Mater. 12 (2000) 1961. [39] I. Nowak, Col. Surf. A 241 (2004) 103. [40] N.D. Petkovich, A. Stein, in: B.-L. Su, C. Sanchez, X.-Y. Yang (Eds.), Hierarchically Structured Porous Materials: from Nanoscience to Catalysis, Separation, Optics, Energy, and Life Science, vol. 1, Wiley-VCH Verlag GmbH and Co. KGaA, Weinheim, Germany, 2012, p. 55. [41] F. Li, Z. Wang, N.S. Ergang, C.A. Fyfe, A. Stein, Langmuir 23 (2007) 3996.