Synthesis of hexagonal mesoporous silicates functionalized with amino groups in the pore channels by a co-condensation approach

Abstract

Mesoporous silicates functionalized with amino groups in the pore channels have been made by the co-condensation of tetraethoxyl siloxide (TEOS) with precursors of P–Si through a triblock copolymer-templated sol–gel process under acidic conditions. Poly(alkylene oxide) block copolymer (P123) was eluted by ethanol extraction and template molecules were removed by refluxing the materials in a mixture of DMSO and water. The resulting materials were characterized in detail by FT-IR, XRD, TEM and N₂ adsorption, in order to study the effect of the precursors on the mesoscopic order and pore structure. Evidence of amino groups located in the pore channels was shown through variation of pore size and BET surface area after the amino groups were coupled with benzaldehyde, and TEM images of materials after staining with RuO₄. Finally, a comparative study of the catalytic performance of materials SBA-Am-10 and SBA-T-10, obtained by two methods, revealed that the catalyst synthesized by our method gave rapid reaction speed and high yields of flavanone by the Claisen–Schmidt condensation of benzaldehyde and 2′-hydroxyacetophenone.

---

1 Introduction

Amino-functionalized mesoporous materials have received considerable attention in recent years among the variety of organo-functionalized mesoporous materials synthesized through the direct synthesis route. They have been found to be effective in base catalyzed reactions^1–5 and further post-synthesis functionalization.^6,7 However, most of the previous work has been on the modification of small MCM-type mesopores,⁸ which were synthesized under basic conditions. Highly ordered mesoporous materials with large pores (>4.5 nm) have recently attracted particular interest from a catalytic perspective. The large pore diameters of the channels facilitate molecular diffusion and access to catalytic sites, and improve the catalytic efficiency. On the other hand, the well-defined channels also have a confining effect on the catalytic selectivity. In several investigations, the confinement of a catalyst in a mesoporous solid improved the activity of the catalyst compared with attachment to amorphous⁹ or microporous silica,¹⁰ due to the enhanced selectivity in a sterically homogeneous environment. It is generally known that antagonistic effects at the mesoscale result in the existence of an optimal catalytic efficiency depending on pore size.¹¹

However, amino-functionalized well-ordered large pore mesoporous materials are perhaps the most difficult to synthesize using a direct synthesis approach in the presence of non-ionic surfactants as structure directing agents under acidic conditions. The prehydrolysis method is probably the most efficient approach to obtain well-ordered materials in which the silica source, usually TEOS or sodium silicate (Na₂SiO₃), is prehydrolyzed for a certain length of time before the functional monomers are added. In order to obtain a well-ordered mesoporous structure, the optimal prehydrolysis time of TEOS was 2 h.¹² In this case, an intact silicate framework was formed before the addition of (3-aminopropyl)triethoxysilane (APTES), which minimized the effect of the protonated amino groups on the self-organization of the surfactants. Protonated species can cross-link with the alkoxysilane precursor during the condensation step, and as a consequence the organosilanes were eventually deposited both in the pore channels and inside the walls of the mesoporous materials.¹³ As a result, amino groups were buried in the silica matrix and on both the inner and outer surfaces of the mesopores.¹⁴ This is not desirable from the point of view of confined catalysis. Another commonly used method to obtain ordered amino-functionalized mesoporous silicates is to use inorganic salts as additives to strengthen the interaction between the silicate species and the surfactant hydrophilic head groups.¹⁵ However, the accessibility of the amino groups in the resulting material remains unknown.

In our previous work on preparing molecularly imprinted mesoporous silicates (MIMS) for 2-naphthol by a surfactant directed sol–gel process, we found that instead of imprinting the template in the matrix, the material turned out to be a periodic mesoporous silicate with amino-functionalized inner mesopores.¹⁶ We rationalized that this was due to the fact that the mono-functionalized organic template tends to interact with the hydrophobic core of the surfactant micelle during the sol–gel process. This directed us to use aromatic groups as protecting groups for the amino groups in order to obtain well-ordered mesoporous materials with amino-functionalized inner pores.

The synthesis of functionalized mesoporous materials through protecting group strategies has been used by several research groups. Schüth’s group synthesized cyano (–CN) functionalized SBA-15 and further hydrolysed them into carboxyl groups (–COOH).¹⁷ Corriu’s group reported the use of N-tert-butyloxycarbamate (-Boc) as a protecting group for preparation of amino functionalized SBA-15.¹⁸ Despite the difficulty in preparing Boc-protected monomers, the ordering of the resulting mesoporous materials was poorly preserved.

The structure and contents of organo-functionalized mesoporous materials synthesized from the co-condensation of TEOS and functional monomers in the presence of a surfactant depends strongly on the nature of the functional monomers. In fact, the strong hydrophobic interaction between phenyl groups and the hydrophobic core of pluronic-type surfactants was theoretically investigated by A. Patti et al. using lattice Monte Carlo simulations. They observed that when the hybrid precursor was sufficiently hydrophobic, it could act as a co-surfactant, swell the core of the surfactant liquid crystal, and lead to structures with smaller interfacial curvature. On the other hand, if the hybrid precursor acted as a co-solvent it would solubilize the surfactant leading to the destruction of the preformed liquid crystal.¹⁹ We envisage that it is possible to obtain highly ordered organo-functionalized mesoporous materials with accessible organic groups in the pore channels by carefully selecting functional monomers with suitable solubility and hydrophobicity.

In this paper, we report the synthesis of amino-functionalized mesoporous materials using several aromatic groups as protecting groups, and compared them with materials synthesized by the prehydrolysis method. The major features of this work are a comparison of the products of both methods regarding the efficiencies of the synthetic routes, precursor effects, the locations of functional groups in the mesoporous products, and their catalytic behaviors for the solid state catalyzed synthesis of flavanone.

2 Experimental

2.1 Chemicals and materials

Phenol (Ph), 2-naphthol (Np), 4-benzylphenol (Bp), 9-anthracenemethanol (Am), di-tert-butyl dicarbonate (DbD) and TEOS were purchased from Gracia Chemical Technology Co., Ltd. Chengdu. 3-Isocyanatopropyltriethoxysilane (IPTES), APTES and pluronic P123 (EO₂₀PO₇₀EO₂₀, MW = 5800 Da) were purchased from Sigma Chemical Co., St. Louis. All other chemicals used were of analytical grade. All solvents and reagents were used without further purification except THF (dried).

2.2 Instrumentation

Infrared spectra were recorded with a Perkin-Elmer system 2000 FT-IR spectrometer. Transmission electron microscopy (TEM) images were obtained on a Tecnai-G2-F20 electron microscope operating at 200 kV. X-ray diffraction (XRD) patterns were obtained at room temperature using an instrument equipped with a Cu Kα X-ray source. The total content of amino groups in the samples was determined by CHN elemental analysis, which was performed using a Perkin-Elmer 2400 Elemental Analyzer. Solid-state cross-polarization (CP) magic-angle spinning (MAS) nuclear magnetic resonance (NMR) analysis was carried out on a Bruker AV-400 spectrometer at 79.49 MHz for ²⁹Si and 100.61 MHz for ¹³C. The ²⁹Si MAS NMR spectra were measured at 60 s repetition delay and 3 μs pulse width. The ¹³C CP MAS NMR spectra were measured at 2 s repetition delay, 2 ms contact time and 2.8 μs ¹H 90° pulse. All chemical shifts were referenced to tetramethylsilane (TMS). Nitrogen adsorption–desorption isotherms were measured at 77 K using a HYA2010-B2 system. The Barrett–Joyner–Halenda (BJH) pore size distribution was calculated from the desorption branch of the isotherm.

2.3 Synthesis of materials

The synthesis of aromatic protecting triethoxysilane precursors (APTPs) was achieved using the method reported in ref. 20. Aromatic moieties (Ph, Np, Bp, Am) were reacted with stoichiometric amounts of IPTES using triethylamine as a catalyst in dried THF at 65 °C until the reaction was complete. The products were obtained by evaporating the solvents, and purified by column chromatography. The products synthesized with Ph, Np, Bp and Am were labelled Ph–Si, Np–Si, Bp–Si and Am–Si, respectively. For comparison, the Boc protecting triethoxysilane precursor synthesized using DbD was obtained according to the method in ref. 18, and labelled Boc–Si.

Amino-functionalized SBA-P-x materials were synthesized by the following procedure:

Throughout this paper, we abbreviate the amino-functionalized mesoporous silica materials as SBA-P-x or SBA-T-x (where P is Ph, Np, Bp, Am or Boc), T denotes APTES, and x is 3, 5, 7, 10, 15 or 20, which is calculated from the molar percentage of [precursor/(precursor + TEOS)]. P–Si denotes triethoxysilane precursors and x denotes the molar percentage of P–Si or APTES in the total silane monomer used in the co-condensation reaction.

P–Si functionalized SBA-15 material was synthesized under acidic conditions from a P123/TEOS/P–Si/HCl/H₂O mixture with a molar composition of 0.0172/1/x%/6/208. Typically, P123 (2.0 g) was fully dissolved in 75 ml of 1.6 M HCl. To this solution was added a stoichiometric amount of P–Si, which was predissolved in 1 ml THF. After the prehydrolysis of P–Si, TEOS (4.45 ml) was added to the mixture and stirred for up to 24 h at 40 °C. The mixture was then transferred into an autoclave for aging at 100 °C for 24 h. After cooling down to room temperature, the resulting particles were isolated by filtration and rinsed with water, and named SBA-P-x-as. P123 was extracted by Soxhlet extraction with ethanol for 24 h, and the remaining material was designated SBA-P-x-PR. By refluxing the particles in a mixture of DMSO and water (v/v, 40/1) at 160 °C for 6 h, the carbamate bond was cleaved and P was removed, except for Boc which was removed by acidic hydrolysis. After filtration, the materials were rinsed with DMSO and ethanol successively, and vacuum dried at 60 °C for 24 h. The prehydrolysis time of P–Si was 1.5 h. Pure SBA-15 was prepared in a similar way but without the addition of functional groups. For comparison, SBA-T-x was synthesized by adding APTES after 2 h of prehydrolysis of TEOS.

2.4 Chemical composition, texture and structural characterization

The nitrogen content, corresponding to the content of incorporated amino groups, was measured by elemental analysis. Texture properties were determined from N₂ adsorption/desorption; specific surface areas were evaluated by the Brunauer–Emmett–Teller (BET) method, and pore size distribution profiles were calculated from the desorption branch of the isotherms using the Barrett–Joyner–Halenda (BJH) method. Information on mesoscopic ordering was obtained by TEM, and crystallographic information was obtained by small angle X-ray diffraction (SXRD).

2.5 Determination of the location of aminopropyl groups

2.5.1 Condensation reaction of amine groups with benzaldehyde. Two kinds of solid amino-functionalized SBA-15 (SBA-Am-10, 0.25 g and SBA-T-10, 0.25 g) were dispersed in anhydrous toluene, and an excess of benzaldehyde (2 equiv. of benzaldehyde per NH₂ moiety) and 100 μl glacial acetic acid were added. The reaction was carried out under reflux for 12 h, and the resulting materials, denoted as SBA-Am-imine and SBA-T-imine, were filtered and washed with an adequate amount of toluene and ethanol to eliminate unreacted benzaldehyde, and after drying at 80 °C for 6 h the changes in pore diameter, BET surface area and pore volumes were determined by nitrogen adsorption measurements.

2.5.2 Characterization by combination of staining technique and TEM. Following a general method¹⁴ for the staining of SBA-15 with RuO₄, SBA-Am-10 and SBA-T-10 were exposed to the vapor of 0.5% RuO₄ (aq.) for 15 min to stain the amino groups. The stained samples were then characterized by transmission electron micrography using a Tecnai-G2-F20 electron microscope working at 200 kV.

2.6 Catalytic reaction

In order to remove residual Cl⁻ ions and neutralize the protonated amine groups, functionalized SBA-15 materials were dispersed in 100 ml of a 0.1 M methanol solution of TMAOH at room temperature for 40 min before the reactions. The resulting solids were recovered by filtration, washed with methanol, and finally dried at 100 °C for 12 h.

2.6.1 Kinetic study of flavanone synthesis catalyzed by SBA-T-10 and SBA-Am-10. Two kinds of amine functionalized SBA-15 (SBA-T-10 and SBA-Am-10) were selected for kinetic study, and the reaction was carried out in a two necked flask equipped with a reflux condenser and a magnetic stirrer bar. Typically, 6 mmol (612 μl) benzaldehyde (A) and 4 mmol (485 μl) 2′-hydroxyacetophenone (B) were stirred in 25 ml DMSO, and 250 mg of SBA-T-10 (or SBA-Am-10) was added into the solution. The mixture was heated to 140 °C with stirring for 14 h. To get the kinetic curve, the reaction was monitored every two hours by withdrawing an aliquot of solution from the flask. The samples were filtered and the filtrates were analyzed using an Agilent HPLC equipped with an Agilent C18 column (4.6 × 250 mm, 5 μm) to determine the yield of flavanone (Fl).

2.6.2 Intramolecular Michael addition of 2′-hydroxychalcone catalyzed by SBA-T-10 and SBA-Am-10. The reaction was carried out in a round-bottom flask equipped with a reflux condenser and a magnetic stirrer bar. Typically, 20 mg SBA-T-10 (or SBA-Am-10) was suspended in 2.0 ml DMSO, and 0.036 mmol (8.0 mg) 2′-hydroxychalcone (Ch) was added to the suspension. The mixture was heated to 140 °C with stirring for 4 h. After cooling to room temperature, the mixture was filtered and the filtrate was analysed by ¹H NMR. The conversion of Ch and the yield of Fl were calculated from the relative volume of C2-H on Fl and β-H of the carbonyl group on Ch. Control experiments were conducted in the presence of blank SBA-15, with and without DMSO, or without added solvent and catalyst under otherwise the same conditions as described above. Each reaction was repeated three times and the average analysed volume vs. time was plotted in Fig. 6.

3 Results and discussion

3.1 Synthesis of materials

The aryl-protected functional monomers were synthesized according to a published procedure.²⁰ Aromatic hydroxyl moieties were reacted with a stoichiometric amount of IPTES using triethylamine as a catalyst in dried THF at 65 °C for 24 h. The products were obtained by the evaporation of the solvents, and were identified as P–Si (Scheme 1(a)). For comparison, the Boc-protected functional monomer (Boc–Si) was prepared following Corriu et al.’s method (Scheme 1(b)).¹⁸

Scheme 1 Synthesis of the aryl-protected functional monomers.

The amino-functionalized mesoporous materials (SBA-P-x) were prepared by the co-condensation of trialkoxysilanes (P–Si) and TEOS through a triblock copolymer-templated sol–gel process¹⁶ with a slight modification. P–Si was added to the surfactant/HCl solution and allowed to prehydrolyse for 1.5 h before the addition of TEOS, to allow the aromatic groups to fully interact with the surfactant micelles. The surfactant was removed by ethanol extraction. To remove the aromatic protecting groups (P), the carbamate bond was cleaved by heating the materials in DMSO/H₂O at 160 °C, while the material synthesized by Boc–Si was heated in an aqueous solution of HCl. After filtration and drying at 60 °C for 24 h, a series of amino-functionalized materials were obtained.

SBA-T-x materials were synthesized by adding APTES after 2 h of prehydrolysis of TEOS.¹² The surfactant was removed by the same method as for SBA-P-x, and the materials were then dried at 60 °C for 24 h for further analysis.

3.2 FT-IR characterization

In order to confirm that P–Si was incorporated into the structure and the successful cleavage of the carbamate bond, the directly synthesized material (SBA-Am-10-as), after surfactant removal (SBA-Am-10-PR) and after protecting group removal (SBA-Am-10) were characterized by Fourier transform infrared spectroscopy (FT-IR, Fig. 1). In all three materials, the clear bands around 1070 and 1220 cm⁻¹ indicate that condensed silica networks are formed (Si–O–Si, asymmetric vibration). For SBA-Am-10-as, absorbance peaks corresponding to C–H stretching and C–H deformation vibrations appear in the range 2850–2900 cm⁻¹, which were due to the CH₂ absorption of P123. This peak is reduced in SBA-Am-10-PR, indicating that the surfactant was almost completely removed upon extraction with ethanol. The characteristic band of the carbamate C=O stretching vibration around 1710 cm⁻¹ is clearly visible for SBA-Am-10-as and SBA-Am-10-PR, indicating the presence of P–Si. After ethanol extraction and DMSO treatment, the 1710 cm⁻¹ peak disappeared (Fig. 1, SBA-Am-10), indicating the cleavage of the carbamate bond. The increasing intensity of the peak around 1640 cm⁻¹ could be assigned to NH₂ bending, which overlapped with the OH bending of absorbed water.²¹ The relatively weak peak at 1503 cm⁻¹ corresponding to symmetric –NH₃⁺ bending²² further proves the successful recovery of the amino groups after removal of the protecting groups.

Fig. 1 FT-IR spectra of SBA-Am-10-as, SBA-Am-10-PR and SBA-Am-10.

3.3 ¹³C and ²⁹Si CP MAS NMR

The incorporation of amino moieties in the mesoporous materials was confirmed by solid-state NMR spectroscopy. The ¹³C CP MAS NMR spectrum of SBA-Am-10-PR (Fig. 2(a)) demonstrates that the Am–Si precursor was incorporated, as shown by the signals at 56.7 ppm (CH₂), 157.1 ppm (carbamate C=O), 124.2–132.6 ppm (anthracene ring), and three additional signals (42.2, 21.3 and 9.1 ppm) which are attributed to the propyl spacer of Am–Si. The removal of protecting groups by treatment with DMSO/H₂O is clearly reflected in Fig. 2(b) by the disappearance of the signals at 56.7, 157.1 and 124.2–132.6 ppm, while the signals of the propyl spacer at 42.7, 21.5 and 9.3 ppm were retained. The ²⁹Si MAS NMR spectrum of SBA-Am-10 (Fig. 3) shows two sets of signals attributed to ternary (T, C–SiO₃, −55 to −68 ppm) and quaternary (Q, SiO₄, −90 to −120 ppm) environments. The presence of T units is further evidence of the existence of CH₂–Si groups. These data could further support the conclusion that aminopropyl moieties were successfully incorporated into the mesoporous material SBA-Am-10, and the protecting groups were successfully removed.

Fig. 2 ¹³C CP MAS NMR spectrum: (a) SBA-Am-10-PR and (b) SBA-Am-10.

Fig. 3 ²⁹Si MAS NMR spectrum of SBA-Am-10.

3.4 Elemental analysis

To further confirm the incorporation of aminopropyl groups, the N content of the samples were analyzed by elemental analysis. The results are presented in Table 1. Percentages of functional monomer (used and integrated) indicate that 81% to 89% of the added functional monomers were integrated in the final materials. We also found that the length of time of functional monomer prehydrolysis significantly affected the N content in the final materials (data not shown in Table 1). The optimal prehydrolysis time was between 1 and 1.5 h. When prehydrolysis time was less than 1 h, only 6–7% of the functional monomers were retained, while prolonging the hydrolysis time above 1.5 h did not increase the amount of functional monomers and resulted in less ordered materials.

Table 1 Incorporated amino contents calculated from elementary analysis

Sample	N content (%)	Amount of amino group (mmol g^−1)	Percentage of functional precursor
			Used^a	Integrated^b
a Molar ratio of P–Si/TEOS as added.b Molar ratio of P/SiO2 calculated using the N content from elemental analysis.
SBA-Np-10	1.36	0.97	10%	8.2%
SBA-Am-10	1.46	1.04	10%	8.8%
SBA-Bp-10	1.34	0.96	10%	8.3%
SBA-Ph-10	1.31	0.93	10%	8.1%
SBA-T-10	1.47	1.05	10%	8.9%
SBA-15	—	—	—	—

---

3.5 Small angle X-ray diffraction measurements

The SXRD patterns of materials prepared from different precursors, SBA-P-10 (P = Ph, Bp, Np, Am, T) are shown in Fig. 4(a), and the effects of the amount of aminopropyl content ranging from 5–20% were demonstrated for SBA-Am-x (x = 5–20%) in Fig. 3(b). The SXRD patterns of the materials in Fig. 4(a) all show an intense peak at d(100) and two weak peaks at d(110) and d(200), which are characteristic of a 2-D hexagonally ordered (P6mm) structure.^23–25 However, all the amino-functionalized materials had a lower degree of ordering than blank SBA-15, which could be attributed to the disturbance of the organic precursors by the ionic interactions between the silicate species and the surfactant hydrophilic head groups.

Fig. 4 XRD patterns of prepared materials (a) SBA-P-5, P = Am, Np, Bp and Ph with the addition of different APTPs, and (b) SBA-Am-x with different concentrations of Am–Si.

The structural ordering is very sensitive to the aromatic groups used in the aromatic group protected precursor P–Si, and the way that P–Si and TEOS are mixed. Well-ordered hexagonal mesoporous materials can be obtained with amino contents up to 20% for SBA-Np-x and SBA-Am-x (only the latter is shown in Fig. 4(b)). The intensity of the reflection declined as the proportion of functional monomer (x) increased, suggesting that less ordered materials were produced. However, for materials prepared with Bp–Si and Ph–Si, the long range ordering was lost when x reached 10%. The peak intensity attributed to the d(100) reflection decreased in the order: SBA-Np-10 > SBA-Am-10- > SBA-Bp-10 > -SBA-T-10 > SBA-Ph-10.

When P–Si was pre-mixed with TEOS before being added into the surfactant solution as is usually done,²⁶ the resulting materials were amorphous. However, when P–Si was added into the surfactant solution 1 h earlier than TEOS, well-ordered hexagonal structures were formed.

The varying intensity of d(100) in Fig. 4(a) may be due to different interaction strengths between the non-polar aromatic groups and the hydrophobic core of the surfactant micelle, which depends on the solubility, size and rigidity of the aromatic moieties.¹⁹ When this hydrophobic interaction is strong enough, it would draw the organic precursors further into the micelles, leading to the formation of well-ordered mesopores (in the case of Np–Si and Am–Si), and resulting in disordered mesopores otherwise (as in the case of Bp–Si and Ph–Si). The addition of organoalkoxysilane into the surfactant solution before TEOS would favor the interaction between the terminal organic moiety of the organoalkoxysilane and the core of the surfactant, thereby promoting the formation of well-ordered mesostructures. Although prehydrolysis of TEOS can also produce well-ordered mesoporous silicates by the formation of an intact silica framework, which could shield the disruptive interactions of the organic monomers with the micelles, the organic monomers were consequently distributed in the silica walls. The reduced intensity of the d(100) peak of SBA-Am-10 relative to that of SBA-T-10 (Fig. 4(a)) could be due to the organic content present in the pores.

It is worth mentioning that when using Boc–Si as precursor, no ordered material was obtained.

3.6 Nitrogen adsorption/desorption characterization

Nitrogen adsorption/desorption isotherms of all obtained materials except SBA-Ph-10 (Fig. 5(a)) exhibited typical type IV isotherms according to IUPAC classification.²⁷ The Brunauer–Emmett–Teller (BET) surface area, pore size, and pore volume are listed in Table 2. The pore sizes of SBA-Bp-10 and SBA-Ph-10 were smaller than those of the blank SBA-15, probably due to the co-solvent effects of Bp and Ph on the surfactant. On the other hand, the pore sizes of SBA-Np-10 and SBA-Am-10 were larger than those of SBA-15, due to the strong interactions between these aromatic groups and the hydrophobic cores of the surfactants (co-surfactant effects), causing the swelling of the hydrophobic cores. Pore size distributions calculated from the desorption branch of the nitrogen adsorption isotherms are presented in Fig. 5(b). Fig. 5(a) shows that pure silica, SBA-Np-10 and SBA-Am-10 show similarly high porosity and well-defined adsorption. The sharp increases in the adsorption at P/P₀ = 0.7–0.8 imply that SBA-Np-10 and SBA-Am-10 possess large pore sizes with narrow distribution. This was conformed further by the pore size distribution measurements (Fig. 5(b)). SBA-T-10 shows a less well-defined hysteresis loop than SBA-Np-10 and SBA-Am-10. SBA-Bp-10 and SBA-Ph-10 display less well-defined capillary condensation steps and lower porosity (Table 2, pore volume). The loss of mesoporosity for SBA-Bp-10 and SBA-Ph-10 might result from the incorporated micelles being more poorly ordered. The decrease in surface area of SBA-Np-10 and SBA-Am-10 compared with SBA-15 (Table 2) is probably due to the strong interactions between the aromatic groups in the hydrophobic–hydrophilic palisade region of the surfactant micelles, which could prevent the hydrophilic head of the surfactants reaching into the pore walls to form micropores.

Fig. 5 (a) Nitrogen adsorption/desorption isotherms of SBA-P-10; (b) BJH pore size distribution of SBA-P-10.

Table 2 Texture data of amine-functionalized SBA-15 prepared with different protecting groups and prehydrolysis

Sample	Nitrogen adsorption
	Pore size (nm)	SBET (m^2 g^−1)	Pore volume (cm^3 g^−1)
SBA-Np-10	7.16	439.625	0.945411
SBA-Am-10	6.58	571.566	0.941257
SBA-Bp-10	6.10	422.182	0.446348
SBA-Ph-10	4.22	498.411	0.464072
SBA-T-10	5.58	545.577	0.760954
SBA-15	6.56	689.486	1.096177

---

The pore sizes of materials made with different aromatic protecting groups are in the order: SBA-Np-10 > SBA-Am-10 > SBA-T-10.

3.7 Transmission electron microscopy (TEM)

The mesoscopic ordering and symmetry inferred from the XRD data can be further confirmed by TEM (Fig. 6). As can been seen, all materials except SBA-Ph-10 (not shown) show excellent hexagonal ordering of mesopores in a uniform longitudinal pore channel, characteristic of SBA-15 materials.

Fig. 6 TEM images of (a) SBA-15, (a′) SBA-15-stained, (b) SBA-Bp-10, (c) SBA-Np-10, (d) SBA-Am-10, (d′) SBA-Am-10-stained, (e) SBA-T-10, (e′) SBA-T-10-stained.

The distributions of the amino groups in materials made by the TEOS prehydrolysis method (SBA-T-10), and through the aromatic protecting group route (SBA-Am-10), were demonstrated by the selective complexation of RuO₄ over amino groups and its subsequent reduction to RuO₂. The sample treatment used was similar to the method reported by Rual Sanz.¹⁴ The high electronic density of this heavy metal makes the passage of electron beams through it difficult when the sample was analysed by TEM. Thus, a specific darkening can be attributed to RuO₂ and, as a consequence, to the position of amino groups in the samples.

The possible interaction of the silanol groups of pure non-functionalized SBA-15 silica with the staining agent was checked as a control. In order to give a better view of the details, we enlarged the magnification of the TEM images for stained SBA-15, SBA-Am-10 and SBA-T-10 (Fig. 6(a′), (d′) and (e′)). Fig. 6(a) and (a′) show the TEM micrographs for SBA-15 before and after the staining treatment. As can be seen, the bright and dark regions correspond to pore cavities and pore walls, respectively. The contrasts are the same in both cases, that is, bright and dark areas correspond to the same regions (pore cavities and pore walls respectively) regardless of the RuO₄ treatment. The results indicate that RuO₄ does not interact with the silanol groups of the silica surface under the staining conditions used in this work.

The TEM image of the stained SBA-T-10 (Fig. 6(e′)) exhibits a number of black spots across the walls, attributed to ruthenium species fixed over amino groups homogeneously distributed in the sample, including pores and walls. This was not the case in SBA-Am-10 (Fig. 6(d′)). The dark areas of RuO₂ were mainly found within the channels of the stained sample, while the pore walls remained bright. This could be evidence of the amino groups being located mostly on the pore surfaces of the mesosilicate. It is also clear that the amino-functionalized samples of SBA-Am-10 and SBA-T-10 present a distribution of pore cavities (bright) and pore walls (dark) similar to that of silica SBA-15 before staining with RuO₄ (Fig. 6(a), (d) and (e)). This could be evidence that amino groups are not detectable by TEM if the staining step is not performed.

3.8 Determination of the location aminopropyl groups by condensation of the amino-functionalized materials with benzaldehyde

To further verify the location of amino groups of materials made by different methods (SBA-T-10 and SBA-Am-10), benzaldehyde (A) was used to form imines with the amino groups in the materials (Scheme 2). If amino groups are located at the pore surface of the mesosilicates, the introduction of A into the pore by formation of an imine with the amino groups at the pore surface would reduce the pore size, or otherwise, the pore size would remain unchanged.

Scheme 2 Condensation reaction of the amine functional group with A.

The mesoporous properties of SBA-T-10 and SBA-Am-10 before and after treatment with A, obtained by nitrogen adsorption experiments, are presented in Table 3. It is interesting that the pore size of SBA-T-10 had reduced slightly from 5.58 to 5.47 nm, while the BET surface area and pore volume of SBA-T-10 did not change significantly after treatment with A. In contrast, the pore size of SBA-Am-10 were reduced from 6.58 to 5.78 nm, and pore volume from 0.941 to 0.773 cm³ g⁻¹. This could be explained by the different locations of the amino groups in SAB-T-10 and SBA-Am-10. Since a TEOS prehydrolysis process was used for the synthesis of SBA-T-10, in which the amino groups were distributed both in the pore channels (minor) and inside the walls of the resulting materials, little A could attach on the pore surface. The changes in pore size, BET surface area and pore volume are negligible. On the other hand, the amino groups are mostly located on the pore surface of SBA-Am-10, so they are available to react with A, resulting in pore size and pore volume changes. These results are consistent with what is observed in the TEM staining images.

Table 3 Texture properties of materials obtained by two methods after treatment with A

Samples	Before reaction			After reaction
	Pore size (nm)	BET surface area (m^2 g^−1)	Pore volume (cm^3 g^−1)	Pore size (nm)	BET surface area (m^2 g^−1)	Pore volume (cm^3 g^−1)
SBA-T-10	5.58	545.6	0.761	5.47	539.8	0.735
SBA-Am-10	6.58	571.6	0.941	5.78	550.8	0.773

---

3.9 Kinetic study of flavanone synthesis

Amino-functionalized SBA-15 type materials prepared by the TEOS prehydrolysis method have been found to be effective base catalysts for the synthesis of flavanone (Fl).^22,28,29 Xueguang Wang et al. found that the yield of Fl and 2′-hydroxyacetophenone (B) catalyzed by amino-functionalized SBA-15 was the highest under solvent free conditions, and DMSO was the best solvent for the reaction. For operational purposes, we used DMSO as a solvent to study the kinetic behavior of the reaction catalyzed by two types of functionalized mesoporous materials. Two materials, SBA-Am-10 and SBA-T-10, with a comparable loading of amine groups based on the elemental analysis (Table 1), were chosen to catalyze the reaction of A and B (Scheme 3). The kinetic behaviors of the reaction were monitored by HPLC analysis.

Scheme 3 Synthesis of Ch and Fl catalysed by amino-functionalized mesoporous silicates.

The total yield of Fl catalysed by SBA-Am-10 was about 20% higher than that of SBA-T-10 (Fig. 7 and Table 4). It also took about 2 h less time to reach the highest yield (Fig. 7). The results could be interpreted as because the amino groups of SBA-Am-10 are more accessible than those of SBA-T-10 for the catalytic synthesis of Fl from A and B.

Fig. 7 Kinetic curves of the yield of Fl from the Claisen–Schmidt condensation of A and B, catalyzed by SBA-Am-10 and SBA-T-10.

Table 4 The catalytic performance of materials in the condensation of A and B with DMSO at 140 °C for 8 h^a

Entry	Catalyst	Conversion of A (%)	Yield of Fl^b (%)
a Reaction conditions: 6 mmol A; 4 mmol B; 2 ml DMSO; 0.25 g SBA-15, SBA-T-10 or SBA-Am-10.b The yield of Fl was analyzed by HPLC.
1	SBA-Am-10	69	44
2	SBA-T-10	53	33
3	SBA-15	<5	<3
4	—	—	—

---

Two steps are involved in the synthesis of Fl: the Claisen–Schmidt condensation of A and B, followed by cyclization of Ch to Fl. To investigate the catalytic performance of the materials, we observed the cyclization of Ch in the presence and absence of catalyst and solvent (Scheme 4). The conversion of Ch in the presence of catalysts, DMSO and no solvent and without materials at 140 °C for 4 h are shown in Table 5.

Table 5 Cyclization of Ch over 20 mg of mesoporous material or without material at 140 °C for 4 h

Entry	Solvent	Catalyst	Yield of Fl (%)
a The yield of Fl was analyzed by NMR. Reaction conditions: 0.036 mmol Ch, 2 ml DMSO.b Solvent free.c Without mesoporous material.d Solvent free and without mesoporous material.
1	DMSO	SBA-Am-10	64^a
2	—	SBA-Am-10	70^b
3	DMSO	SBA-T-10	62^a
4	—	SBA-T-10	71^b
5	DMSO	SBA-15	69^a
6	—	SBA-15	61^b
7	DMSO	—	66^c
8	—	—	9^d

---

As can been seen from Fig. 8, the two curves almost overlap, indicating that both of the materials are efficient in terms of converting Ch into Fl. In other words, the cyclization is not related to the distribution of amino groups in the materials.

Fig. 8 Kinetic conversion of Ch to Fl.

To verify if the mesopores were responsible for the cyclization reaction, blank SBA-15 with/without DMSO was used for the same reaction. The selectivity for Fl was also similar to that of the amino-functionalized materials. Surprisingly, when no solid catalyst was added to a DMSO solution of Ch, the selectivity for Fl was 66%. When Ch alone was heated to 140 °C for 4 h, only 9% of Fl was detected. The results reveal that DMSO, amino-functionalized SBA-15 and non-functionalized SBA-15 can catalyze the cyclization of Ch. Comparing the data in Tables 4 and 5, we could conclude that amino-functionalized SBA-15 was an effective catalyst for the Claisen–Schmidt condensation. Mesoporous silicates made by the aryl-protecting group method are more efficient for the Claisen–Schmidt condensation than that made by the conventional TEOS prehydrolysis method²⁸ (Scheme 4).

Scheme 4 Reaction of Ch.

4 Conclusions

A synthetic method for hexagonally-ordered mesoporous silicates functionalized by amino groups in the pore channels was developed by an aromatic protecting group approach. Four aromatic groups were used and it was found that the solubility and hydrophobicity of the aromatic groups affected the ordering of the resulting materials. The most critical difference between SBA-Am-10 and SBA-T-10 was the distribution of the amino groups integrated in the materials, based on TEM and A condensation experiments. The advantage of the materials in the catalysis of Fl synthesis were compared with materials made by the TEOS prehydrolysis method. A faster reaction and higher yield of Fl was obtained with the SBA-Am-10 catalyzed Claisen–Schmidt condensation of A and B. Further investigation of the mesoporous silicates with other reactions is under way in our laboratory.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (NSFC 21442012 and NSFC 21372213).

Notes and references

1. T. Yokoi, H. Yoshitake, T. Yamada, Y. Kubota and T. Tatsumi, J. Mater. Chem., 2006, 16, 1125.
2. Y. Xia and R. Mokaya, Angew. Chem., Int. Ed., 2003, 42, 2639.
3. J. Mdoe, J. H. Clark and D. J. Macquarrie, Synlett, 1998, 6, 625.
4. L. Martins, W. Hölderich, P. Hammer and D. Cardoso, J. Catal., 2010, 271, 170.
5. T. M. Suzuki, T. Nakamura, K. Fukumoto, M. Yamamoto, Y. Akimoto and K. Yano, J. Mol. Catal. A: Chem., 2008, 280, 224.
6. A. C. Blanc, D. J. Macquarrie, S. Valle, G. Renard, C. R. Quinn and D. Brunel, Green Chem., 2000, 2, 283.
7. A. S. M. Chong and X. S. Zhao, Catal. Today, 2004, 93–95, 293.
8. D. J. Macquarrie and D. B. Jackson, Chem. Commun., 1997, 1781.
9. M. J. Climent, A. Corma, S. Iborra and J. Primo, J. Catal., 1995, 151, 60.
10. A. Corma, Chem. Rev., 1997, 97, 2373.
11. F. Goettmann and C. Sanchez, J. Mater. Chem., 2007, 17, 24.
12. Q. Wei, Z. R. Nie, Y. L. Hao, L. Liu, Z. X. Chen and J. X. Zhou, J. Sol-Gel Sci. Technol., 2006, 39, 103.
13. A. S. M. Chong, X. S. Zhao, A. T. Kustedjo and S. Z. Qiao, Microporous Mesoporous Mater., 2004, 72, 33.
14. R. Sanz, G. Calleja, A. Arencibia and E. S. Sanz-Pérez, Microporous Mesoporous Mater., 2012, 158, 309.
15. C. Yu, B. Tian, J. Fan, G. D. Stucky and D. Zhao, Chem. Commun., 2001, 2726.
16. G. T. Luo, Y. P. Li, A. Wang, Q. Lin, G. L. Zhang and C. Wang, Open Chem., 2015, 13, 756.
17. C. M. Yang, B. Zibrowius and F. Schüth, Chem. Commun., 2003, 1772.
18. A. Mehdi, C. Reyé, S. Brandès, R. Guilard and R. J. P. Corriu, New J. Chem., 2005, 29, 965.
19. A. Patti, A. D. Mackie, V. Zelenak and F. R. Siperstein, J. Mater. Chem., 2009, 19, 724.
20. J. E. Lofgreen, I. L. Moudrakovski and G. A. Ozin, ACS Nano, 2011, 5, 2277.
21. N. T. S. Phan and C. W. Jones, J. Mol. Catal. A: Chem., 2006, 253, 123.
22. X. G. Wang, S. K. Kyle, J. C. C. Chan and S. Cheng, J. Phys. Chem. B, 2005, 109, 1763.
23. C. T. Kresge, M. E. Leonowicz, W. J. Roth, J. C. Vartuli and J. S. Beck, Nature, 1992, 359, 710.
24. G. S. Attard, J. C. Glyde and C. G. Göltner, Nature, 1995, 378, 366.
25. D. Y. Zhao, Q. S. Huo, J. L. Feng, B. F. Chmelka and G. D. Stucky, J. Am. Chem. Soc., 1998, 120, 6024.
26. Y. J. Gu and B. Yan, Eur. J. Inorg. Chem., 2013, 2963.
27. K. S. Sing, W. D. H. Everett, R. A. W. Haul, L. Moscou, R. A. Pierotti, J. Rouquérol and T. Siemieniewska, Pure Appl. Chem., 1985, 57, 603.
28. X. G. Wang, J. C. C. Chan, Y. H. Tseng and S. Cheng, Microporous Mesoporous Mater., 2005, 85, 241.
29. S. Y. Chen, C. Y. Huang, T. Yokoi, C. Y. Tang, S. J. Huang, J. J. Lee, J. C. C. Chan, T. Tatsumi and S. Cheng, J. Mater. Chem., 2012, 22, 2233.

---

Footnote

† These authors contributed equally to this work.

---