Synthesis, characterization and catalytic activity of ordered SBA-15 materials containing high loading of diamine functional groups

(1)

Synthesis, characterization and catalytic activity of ordered

SBA-15 materials containing high loading of diamine functional groups

Xueguang Wang, Jerry C.C. Chan, Yao-Hung Tseng, Sooﬁn Cheng

*

Department of Chemistry, National Taiwan University, Taipei 106, Taiwan Received 4 November 2005; received in revised form 3 May 2006; accepted 5 May 2006

Available online 21 June 2006

Abstract

Diamine-functionalized mesoporous SBA-15 materials have been synthesized by co-condensation of tetraethylorthosilicate (TEOS) with [3-(2-aminoethyl aminopropyl)]trimethoxysilane (ATMS) using amphiphilic block copolymer P123 as pore-directing agent under the aid of inorganic salt NaCl and with prehydrolysis of TEOS prior to the addition of organosiloxane. The functionalized materials have hexagonal mesostructure ordering, narrow pore-size distribution with diameter around 60 A˚ , and high loadings of amino groups (up to 3.49 mmol g1). Prehydrolysis of TEOS was a key step for preparation of well-ordered mesoporous materials with amino func-tionalities. Inorganic salt had great inﬂuence on the ordering of mesostructure and the morphology of the resultant materials. The dia-mine-functionalized materials showed better catalytic performance than the aminopropyl-functionalized counterpart in the Knoevenagel reaction of benzaldehyde with ethyl cynoacetate to form a,b-unsaturated compound in liquid phase.

Keywords: SBA-15; Aminoethylaminopropyl; Mesostructure; Co-condensation; Functionalization; Knoevenagel reaction

1. Introduction

Since the discovery of the M41S class of mesoporous sil-icates[1], incorporation of organic groups to impart func-tionalities to the pore surface of the mesoporous silica materials has attracted much attention due to the potential applications of the resultant materials in the ﬁeld of catal-ysis [2–5], separation [6,7], sensor design [8], and nano-science [9,10]. There are two approaches for surface functionalization, i.e. grafting (also known as post-synthe-sis) and direct synthesis or co-condensation[11]. In post-synthesis, organic functional groups are grafted through the reaction of a silane coupling agent with the free and germinal silanol groups on the surface of mesopores. The organic functional groups can be easily chosen and designed to meet diﬀerent requirements. The resultant materials generally maintain highly ordered structures

and show relatively high hydrothermal stability after graft-ing reaction [12]. However, the distribution of the func-tional groups on the surface of the pore wall is likely not uniform and the organic groups are grafted mainly on the external surface of the mesoporous particles or near the pore mouth because of the mass transfer[13]. Compar-atively, the direct synthesis pathway by co-condensation of siloxane and organosiloxane precursors often oﬀers a bet-ter control of the resultant mabet-terials in bet-terms of a higher and more uniform surface coverage of the organic functionalities without the blockage of the mesopores

[14]. Nevertheless, the resultant materials usually show less structural ordering than the pure siliceous counterpart, and the organosiloxane precursors must be carefully chosen to avoid possible phase separation and Si–C bond cleavage during the synthesis and surfactant removal processes

[10]. Up to date, just several of the organic functional groups can be synthesized through the direct co-condensa-tion method.

There has been considerable interest in the development of heterogeneous solid catalysts, since the use of heterogeneous

*

Corresponding author. Tel.: +886 2 23638017; fax: +886 2 23636359. E-mail address:[email protected](S. Cheng).

(2)

catalytic processes allows easier separation, recovery and recycling of the catalyst from the reaction mixture. More-over, heterogeneous catalysts sometimes gave better selec-tivities than the homogeneous ones in many bimolecular reactions [15]. Amino-functionalized mesoporous silicas have been widely investigated due to the applications in base-catalyzed reactions [14,16]. However, most of the work was on the modiﬁcation of MCM-type [17,18], and HMS-type [19] materials with relatively small mesopore sizes, which were synthesized under basic or neutral condi-tions. In the large mesoporous silica system, disordered materials were often obtained by direct co-condensation of tetraethylorthosilicate (TEOS) and amino-functional silane due to the interference of the protonated amine groups in the self-assembly of the copolymer surfactant and the silica precursor under the strong acidic condition

[20,21]. However, functionalized mesoporous materials with large pores are desired in many applications such as the immobilization and encapsulation of large molecules

[21,22]. Recently, we have successfully synthesized large-pore ordered SBA-15 silica functionalized with a high loading of aminopropyl groups via prehydrolysis of TEOS prior to the addition of APTES[23,24]. Moreover, in the attempt to incorporate more hydrophobic functional groups such as N-methylaminopropyl groups, the addition of inorganic salts in the synthesis mixture was found to signiﬁcantly improve the crystallographic ordering of the mesoporous silica [25]. The salt eﬀect in the synthesis of mesoporous materials has been reported to improve the hydrothermal stability [26], control the morphology [27], tailor the framework porosity[28], and improve the meso-structure ordering[29].

Diamine-functionalized mesoporous materials are

attracting much attention due to the strong absorbing and chelating performance of ethylenediamine (ED) groups. Dai et al. [6] synthesized diamine-functionalized ordered mesoporous materials to selectively absorb the tar-geted metal ions based on the aﬃnity of the surface-coated functional ligand for a speciﬁc metal ion. Zhang et al.[30]

successfully prepared transition metal oxide or sulﬁde nanoparticles in ordered mesoporous silicas with ethylene-diamine groups by post-synthesis. More recently, con-trolled Pt and Au nanoparticles in ordered mesoporous materials have also been reported using diamine groups on the pore surface as stabilizer[31–33]. Even more inter-estingly, mesoporous ethane-silicas functionalized with dia-minocyclohexane showed a high catalytic activity and the enantioselectivity in the asymmetric transfer hydrogenation of acetophenone [34]. These urges us to report large-pore ordered silica containing diamine groups using triblock copolymer Pluronic P123 (EO20PO70EO20) as

pore-direct-ing agent under a strong acidic condition through the com-bination of TEOS prehydrolysis and inorganic salt, which is expected to have potential applications in the prepara-tion of controlled high loading metal or metal oxide nano-particles in the mesoporous silica and to have better catalytic activity in solid base catalytic reactions.

2. Experimental

Surfactant P123 (EO20PO70EO20, Mav 5800) and

[3-(2-aminoethyl)aminopropyl] trimethoxysilane (ATMS,

97%) were purchased from Aldrich. Other chemicals were from Acros. All chemicals were used as received.

2.1. Synthesis of materials

Diamine-functionalized SBA-15 materials were pre-pared through co-condensation method. In the typical syn-thesis, 4 g of Pluronic 123 and 11 g of NaCl (1.5 M) were dissolved in 125 g of 2.0 M HCl solution with stirring at room temperature. After adding TEOS, the resultant solu-tion was hydrolyzed at 40C for 0 or 2 h before ATMS was slowly added in. The molar composition of the mixture was (1 x)TEOS:xATMS:6.1HCl:4.6NaCl:0.017P123:165H2O,

where x varied from 0 to 0.20, or ATMS/(TEOS + ATMS) molar ratio being 0–20%. The resulting mixture was stirred at 40C for 20 h and then transferred into a polypropylene bottle and aged at 90C under static condition for 24 h. The solid product was recovered by filtration and dried at room temperature overnight. The resultant samples are denoted as SBA-xED-P-S, where x is the ATMS/(TEOS + ATMS) molar ratio, P represents pre-hydrolysis of TEOS, and S represents the addition of NaCl salt. The template was removed from the as-synthesized material by refluxing in 95% ethanol for 24 h. Finally, the material was filtered, washed several times with water and ethanol and dried at 50C. A functionalized sample containing 10% ATMS with TEOS prehydrolysis in the absence of NaCl was denoted as SBA-10ED-P.

For comparison, 10 mol% aminopropyl-functionalized SBA-15 (denoted as SBA-10NH2) was prepared according

to literature[23]. Four grams of Pluronic 123 was dissolved in 125 g of 2.0 M HCl solution at room temperature. After prehydrolysis of TEOS (7.84 g) for 2 h, aminopropyltrieth-oxysilane (APTES) was slowly added into the solution. The resulting mixture was stirred at 40C for 20 h and then transferred into a polypropylene bottle and reacted at

90C under static condition for 24 h. APTES/(TEOS

+ APTES) = 0.1. The template was removed by reﬂuxing in ethanol.

2.2. Sample characterization

X-ray powder diﬀraction (XRD) patterns were obtained on a PANalytical X’Pert Pro diﬀractometer using Cu Ka radiation (k = 1.5418 A˚ ) at 45 kV and 40 mA. The data were collected from 0.5 to 5 (2h) with a resolution of 0.02.

N2 adsorption–desorption isotherms were measured

using Micromeritics Tristar 3000 at liquid nitrogen temper-ature. Before the measurements, the samples were degassed

at 100C overnight. The speciﬁc surface areas were

evaluated using Brunauer–Emmett–Teller (BET) method. Pore size distribution (PSD) was calculated using the

(3)

Barrett–Joyner–Halenda (BJH) method based on the adsorption branch of the isotherms, and the pore size was reported from the peak position of the distribution curve. The pore volume was taken at the P/P0= 0.990

point.

Thermogravimetric (TG) analyses were carried out on a Du Pont 951 thermogravimetric analyzer with a heating speed of 10C/min under air in a ﬂow of 50 ml/min. N ele-mental analyses (EA) were performed on a Heraeus CHNS elemental analyzer. Fourier transform infrared (FTIR) was carried on a Nicolet Magna-IR 550 Spectrometer with a resolution of 2 cm1 using the KBr method.

Transmission electron microscopy (TEM) was per-formed on a Hitachi H-7100 electron microscope, operat-ing at 75 kV. The scannoperat-ing electron microscopy was carried out on a Hitachi S-800 Electron Microscope.

The29Si,13C and1H NMR experiments were carried out at frequencies of 59.6, 75.5 and 300.1 MHz, respectively, on a Bruker DSX300 NMR spectrometer equipped with a commercial 7 mm MAS NMR probe. All spectra were measured at room temperature. The magic-angle spinning frequencies were set at 6 kHz for all experiments and the variation was limited to ±3 Hz using a commercial pneu-matic control unit. Chemical shifts were externally refer-enced to TMS for29Si and 13C.

2.3. Catalytic reactions

Before the reaction, the functionalized SBA-15 materials were treated with methanol solution of tetramethylamo-nium hydroxide (TMAOH) to remove the residue Clions and to neutralize the protonated amine groups. 1 g of the ethanol-extracted sample was suspended in 50 ml of 0.2 M methanol solution of TMAOH at room temperature for 20 min. The solid was recovered by filtration, washed with methanol, and finally dried at 120C for one day. All the catalytic reactions were carried out in a flask with a magnetic stirrer immersed in a thermostat bath. In a typ-ical experiment, 10 mmol (1.15 g) of ethyl cynoacetate and 10 mmol (1.08 g) of benzaldehyde were mixed in cyclohex-ane and kept at 35C, and then 0.15 g of the dried catalyst was rapidly added into the reactor. After the reaction, the catalyst was separated by filtration. The products were analyzed using a Chrompak CP 9000 gas chromatograph (GC) equipped with 30 m· 0.32 mm RTX-50 capillary col-umn and FID detector. 0.2 g of decane was used as internal standard. Individual reaction product was identified by GC–Mass spectrometry (HP6890 mass spectrometer con-nected with a 30· 0.25 mm RTX-50 capillary column). 3. Results and discussion

3.1. Synthesis and characterization of mesoporous materials The functionalized materials were ﬁrst prepared through one-pot synthesis without TEOS prehydrolysis. No precip-itate was observed in the mixture of TEOS and ATMS at

40C under acid condition. A transparent gel was formed in subsequent aging at 90C in case of 5% ATMS mixture. When the ATMS molar content was increased to 10%, no gel or precipitation was observed even in the aging period. On the contrary, the sample synthesized with TEOS prehy-drolysis such as SBA-10ED-P shows an apparent diﬀrac-tion peak in the XRD pattern with a d-spacing of 88 A˚ , indexed to the (1 0 0) plane of hexagonal phase. These results indicate that TEOS prehydrolysis prior to the addi-tion of ATMS is a key step for the successful synthesis of ordered diamine-functionalized SBA-15. Similar results were also observed in the synthesis of other organic-func-tionalized SBA-15 [11,24]. This feature was explained by that organic functional groups could disturb the assembly of the mesophase [35]. In the present case, the protonated diamine groups probably interact with the ethoxy groups of TEOS strongly through hydrogen bond so that the hydrolysis and condensation of ethoxysilane are inhibited

[24]. Through prehydrolysis of TEOS, the surfactant micelles could assemble with TEOS without the perturba-tion from the organic funcperturba-tional groups.

It was reported that the addition of inorganic salt has great influence on the hydrolysis, condensation and aggre-gation kinetics of the silica precursors [36]. After addition of 1.5 M NaCl, the SBA-10ED-P-S sample with 2 h prehy-drolysis of TEOS shows a more intensive (1 0 0) reflection and two more well-resolved peaks indexed to (1 1 0) and (2 0 0) reflections, respectively, in comparison with SBA-10ED-P synthesized without salt (Fig. 1). The d100-spacing

also increases from 88 A˚ to 91 A˚ (Table 1). On the other hand, for the sample synthesized with 1.5 M NaCl but without TEOS prehydrolysis, only X-ray amorphorous precipitate was obtained with very low surface area as seen in Fig. 1 and Table 1. These results further conﬁrm that TEOS prehydrolysis prior to the addition of ATMS is essential in synthesis of ordered diamine-functionalized SBA-15.

The ATMS concentration in the reaction mixture has inﬂuence on the mesopore ordering of the materials. The

XRD patterns in Fig. 1 of the materials with ATMS/

(TEOS + ATMS) molar ratio less than 0.15 and synthe-sized with NaCl show one intense (1 0 0) reflection and two well-resolved weak reflections of (1 1 0) and (2 0 0) planes, respectively. As the ATMS molar ratio increases to 0.20, only one weak (1 0 0) reflection was observed, showing that high concentration of organic groups would reduce the mesostructure ordering. The d100-spacing of

the materials increases slightly with the increase in the ATMS content, except SBA-20ED-P-S which has low mes-ophase ordering. This variation in d100-spacings is diﬀerent

from previous observation on the functionalized MCM-41 with triethoxyvinylsilane [37] and functionalized SBA-15 with 3-aminopropyltriethoxysilane [23]. It is generally accepted that the cell dimension of organic-functionalized mesoporous materials decreases with organic content because the strong hydrophobic interaction between the non-polar organic groups and the hydrophobic portion

(4)

of the surfactant molecules would draw the organic precur-sors into the micelles and lead to a decrease in the d100

-spacing [17]. In the diamine-functionalized SBA-15, the hydrophobic interactions between the organic moiety of ATMS and the hydrophobic PO portion of the surfactant P123 are probably interrupted by the hydrophilic diamine groups. The larger d100-spacing observed with higher

ATMS content is probably due to more hydrated water molecules are present in the interface of surfactant and silicate framework. In comparison with the as-synthesized sample, the (1 0 0) reﬂection peaks of the extracted materi-als shift slightly toward larger 2h angles, indicating a slight shrinkage of the cell dimension upon extraction.

TEM images of the extracted functionalized materials are illustrated in Fig. 2. For diamine-functionalized SBA-15 synthesized without NaCl in the reaction mixture, only those with less than 10 mol% ATMS could have clear one-dimensional channels been seen (Fig. 2(a)). In contrast, all the functionalized SBA-15 materials containing up to 20 mol% ATMS displays well-ordered one-dimensional pore structure when NaCl was added in the synthesis mix-tures. These results further conﬁrm that inorganic salt can improve the mesostructure ordering. It is also well known that inorganic salt in the reaction mixture could inﬂuence

the morphology of the mesoporous materials. Fig. 3

compares the SEM micrographs of SBA-10ED-P and SBA-10-P-S. It can be seen that SBA-10ED-P synthesized without NaCl consists of short rod-shaped particles of 600 nm in length and 300 nm in diameter, while SBA-10ED-P-S appears to be ﬁber-like material with the length as long as several tens of micrometers. Moreover, these ﬁbers of900 nm in diameter are aggregated to form bun-dles of6 lm in diameter.

Thermal stability of the organic functionalized material was performed by TG analysis in air on the as-synthesized and ethanol-extracted SBA-10ED-P-S, and the results are shown in Fig. 4. Both samples show ﬁrst weight loss of 4–7% at temperatures lower than 100C, corresponding to the desorption of physically adsorbed water or ethanol. This is followed by a large weight loss of34% in the tem-perature range of 150–500C for the as-synthesized sample (Fig. 4(a)). The corresponding DTG proﬁle displays two peaks centered at 254C and 347 C. The former peak is due to the thermal removal of P123 surfactant and the lat-ter is mainly attributed to the decomposition of organic functional groups. For the sample after ethanol extraction, only one weight loss of ca.10% was observed centered at 274C, which is attributed to the decomposition of organic

1 2 3 4 5 1 2 3 4 5 SBA-20ED-P-S SBA-10ED-P-S SBA-15ED-P-S SBA-5ED-P-S SBA-10ED-P SBA-10ED-S In te n s ity (a.u .) 2θ / o (100) _(b) (100) SBA-20ED-P-S SBA-15ED-P-S SBA-10ED-P-S SBA-5ED-P-S SBA-10ED-P SBA-10ED-S Intensity (a.u.) 2θ / o (a)

Fig. 1. XRD patterns of (a) as-synthesized and (b) extracted SBA-15 materials with diﬀerent ATMS contents in the reaction mixture.

Table 1

Physico-chemical properties of the extracted functionalised SBA-15 materials containing diﬀerent concentrations of ATMS

Sample d100spacing/A˚ Pore diameter/A˚ SBET/m2g1 Pore volume/cm3g1 Wall thicknessa/A˚ N contentb

SBA-10ED-S – – 1.6 0.003 – 2.61 (2.89) SBA-10ED-P 88 58 316 0.36 44 2.17 (2.89) SBA-5ED-P-S 90 67 476 0.60 37 1.05 (1.55) SBA-10ED-P-S 91 63 375 0.48 42 2.13 (2.89) SBA-15ED-P-S 92 60 265 0.33 46 2.86 (4.06) SBA-20ED-P-S 91 54 37 0.04 51 3.49 (5.09) SBA-10NH2 92 67 664 0.85 40 1.21 (1.54) a

Calculated by aopore size (ao= 2d100/

p 3).

b

(5)

groups incorporated in materials. These results indicate that the P123 surfactant has been eﬀectively removed from the as-synthesized sample through ethanol extraction. When P123 was used as the template for pure siliceous SBA-15, the block copolymer species was decomposed at around 145C[38]. However, in this system, the decompo-sition temperature of P123 surfactant is apparently ele-vated. On the other hand, the decomposition temperature

of the organic functional groups decreased greatly from 347C to 274 C after the template was extracted. These results imply that the amine groups might have a strong interaction with the surfactant. The 1–3% weight losses above 500C are likely due to the dehydroxylation of the silicate networks[39].

The incorporation of organic functional groups in the silica framework was also conﬁrmed by FTIR spectra as

Fig. 2. TEM images of the solvent-extracted diamine-functionalized 15: (a) 10ED-P, (b) 10ED-P-S, (c) 15ED-P-S, and (d) SBA-20ED-P-S.

(6)

shown in Fig. 5. The typical Si–O–Si bands around 1220, 1070, 795, and 470 cm1associated with the condensed sil-ica network are observed on the as-synthesized and all the extracted samples. For the functionalized samples, a weak peak around 690 cm1, attributed to N–H bending vibra-tion, and another weak peak at 1510 cm1 corresponding to the symmetric NHþ3 bending vibration can be seen,

showing the incorporation of amino groups. In addition, the strong peak around 1635 cm1 is mainly assigned to N–H bending vibration, which is overlapped with the bending vibration of adsorbed H2O. The absorbance peak

of the C–N stretching vibration in the wavenumber range of 1000–1200 cm1 cannot generally be resolved due to its overlap with the absorbance of Si–O–Si stretch in the range 1000–1130 cm1 and that of Si–CH2–R stretch in

the range 1200–1250 cm1 [21]. The weak bands around 1450–1480 cm1, associated with –CH2 vibrations can be

seen for the functionalized materials and the intensity increases with the ATMS content in the reaction mixture. This further conﬁrms the incorporation of organic species in the framework. Similarly, the absorbance peaks associ-ated with non-condensed Si–OH groups in the range 940–970 cm1can also be observed and the peak position gradually shift toward lower wavenumber from 961 cm1 to 942 cm1 with the increase of ATMS concentration in

the reaction mixture, similar to that observed on amino-propyl-functionalized SBA-15 [24]. This can be explained by the increase in interaction between the amine groups and the silanol groups through hydrogen bonding [16]. The presence of organic groups was further corroborated by a broad band in the range 2700–3400 cm1and by the increase of its intensity with diamine content. The absor-bances in the range 3000–2700 cm1are due to the stretch-ing of –CH2groups and those in the range 3000–3400 cm1

are assigned to the asymmetric and symmetric stretching vibrations of NH2 groups. The broadening of the silanol

bands around 3480 cm1, may be due to the cross-linking through the hydrogen-bonding interaction between the amine groups and the silanol groups. In comparison with the as-synthesized sample, the absorbance peaks associated with P123 surfactant disappear in the spectra of the extracted samples, further conﬁrming the removal of the surfactant by ethanol extraction.

Solid state 13C and 29Si NMR spectroscopies were proved to be the most useful tool for providing chemical information with respect to the condensation of organo-siloxane.29Si MAS NMR spectrum of the extracted SBA-10ED-P-S in Fig. 6(a) displays three distinct resonance peaks at upper ﬁeld corresponding to Q4(d =110 ppm), Q3(d =101 ppm) and Q2

(d =91 ppm), and two weak peaks at lower ﬁeld, assigned to T3(d =66 ppm) and T2 (d =58 ppm), respectively, where Qn= Si(OSi)n(OH)4n,

n = 2–4 and Tm= RSi(OSi)m(OH)3m, m = 1–3. The

appearance of Tmpeaks conﬁrms that the organosiloxane precursor is condensed as part of the silicate framework. The relative integrated intensity of the Tmand Qnsignals (Tm/(Tm+ Qn)) is 0.090, which is close to the expected value of 0.10 on the basis of the composition of AEAPTMS in the reaction mixture. The13C CP-MAS NMR spectrum of SBA-1-ED-P-S inFig. 6(b) clearly shows ﬁve resonance peaks at 10, 21, 38, 47 and 52 ppm, corresponding to the C atoms on the Si–CH2–CH2–CH2–NH–CH2–CH2–

NH2 group in sequence from left to right, respectively

[18]. This result further conﬁrms that ATMS precursors

100 200 300 400 500 600 700 800 60 70 80 90 100 Temperature / oC Weight loss (%) (a) 254oC 347oC -0.4 -0.2 0.0 0.2 DTG 100 200 300 400 500 600 700 800 70 80 90 100 Temperature / oC Weight loss (%) (b) 274oC -0.4 -0.2 0.0 0.2 DT G

Fig. 4. TGA and DTG proﬁles of (a) the as-synthesized and (b) the extracted SBA-10ED-P-S. 4000 3500 3000 2500 2000 1500 1000 500 e d b Transmittance (%) Wavenumber (cm-1) a c f NH+ NH+ +H₂O NH+ C-N Si-OH NH P123

Fig. 5. FTIR spectra of (a) as-synthesized SBA-10ED-P-S and the ethanol extracted SBA-ED-P-S with diﬀerent concentrations of ATMS synthesized in the presence of NaCl: (b) 0%, (c) 5%, (d) 10%, (e) 15%, and (f) 20%.

(7)

are co-condensed into the mesoporous silca and the organic moieties are not decomposed during the preparation proce-dure. Besides, no resonance peaks corresponding to the surfactant P123 were observed in the range of 67–77 ppm, indicating that the surfactant was completely removed during the extraction as also supported by TG and IR spectra.

Chemical elemental analysis of nitrogen was used to quantitatively determine the incorporation of amine func-tional groups. Based on the results inTable 1, about 70% of the ATMS precursor was eﬀectively incorporated in the silica framework. The incorporation percentages are lower than80% for aminopropyl-functionalized SBA-15

[24], implying that the diamine groups are more seriously interfering the assembly of mesostructure than the single amine groups. As a result, the N content of SBA-10ED-P-S is lower than twice the amount of SBA-10NH2.

Never-theless, the N content could reach as high as 3.49 mmol g1 when the ATMS/(TEOS + ATMS) molar ratio in the ini-tial mixture was 0.2, although the mesostructure ordering was not good and the surface area was low for the material with such a high N content. In comparison of the materials synthesized with and without TEOS prehydrolysis (SBA-10ED-S versus SBA-10ED-P-S), it is noticed that the amine content decreases with TEOS prehydrolysis. In contrast, the addition of inorganic salt does not aﬀect the amine con-tent signiﬁcantly.

N2 adsorption–desorption isotherms of the extracted

functionalized materials are illustrated in Fig. 7(a). The SBA-10ED-S sample synthesized without TEOS

prehy-drolsis shows negligible amount of nitrogen adsorption. All the other samples with TEOS prehydrolysis containing ATMS/(TEOS + ATMS) molar ratio less than 0.15 dis-plays type IV isotherms with steep increases in adsorption at relative pressure P/P0around 0.6–0.8, similar to that of

pure SBA-15. As expected for organic-functionalized mes-oporous silica, the BET surface area, BJH pore sizes and pore volumes of the materials decrease gradually with ATMS content, due to the occupation of large organic groups in the pore channels. On the contrary, the wall thickness of the materials increases with ATMS content. BJH pore size analysis shows that the materials synthesized with TEOS prehydrolysis and the aid of NaCl contain mes-opores of narrow pore size distribution (PSD) with diame-ters in 60–67 A˚ range. In contrast, sample SBA-10ED-P which was synthesized without NaCl shows a broader PSD and smaller pore diameter (Fig. 7(b)). Therefore, the inorganic salt NaCl not only inﬂuences the mesostructure ordering and the morphology of the materials, but also enhances the BET surface area, pore size and pore volume. When the ATMS concentration further increases to a ratio of 0.20, a very small hysteresis loop was observed, and that was attributed to poor assembly of the mesostructure in high ATMS concentration.

120 100 80 60 40 20 0 -20 -40 -60 C3 C5 C4 _C2 C1 5 4 3 2 1 Si-CH₂-CH₂-CH₂-NH-CH₂-CH₂-NH₂ Chemical shift / ppm (b) -50 -100 -150 -200 T2 T3 Q2 Q3 Q4 Chemical shift / ppm T/(Q+T) = 0.090 (a)

Fig. 6. (a)29Si MAS NMR spectrum and (b)13C {1H} CPMAS NMR spectrum of the extracted SBA-15 with 10% ATMS synthesized in the presence of NaCl. 0.0 0.2 0.4 0.6 0.8 1.0 0 100 200 300 400 Volume adsorbed cm 3 / g, STP Relative pressure (P/P0) SBA-10ED-S SBA-10ED-P SBA-5ED-P-S SBA-10ED-P-S SBA-15ED-P-S SBA-20ED-P-S (a) 50 100 150 200 0.00 0.01 0.02 0.03 0.04 0.05

Pore size distribution (cm

3 g -1 A -1 )

Pore size (A)

SBA-10ED-P SBA-5ED-P-S SBA-10ED-P-S SBA-15ED-P-S SBA-20ED-P-S (b)

Fig. 7. (a) Nitrogen adsorption–desorption isotherms and (b) BJH pore size distribution plots of the extracted functionalized materials with diﬀerent ATMS contents in the reaction mixture.

(8)

3.2. Catalysis reactions

The Knoevenagel condensation is of great importance to the synthetic chemists in the construction of new C–C bonds[40]. Amine-functionalized silica has been found to be eﬀective base catalysts for this synthesis [14]. Here we choose the Knoevenagel reaction of benzaldehyde with ethyl cynoacetate (Scheme 1) to form a,b-unsaturated com-pound (2-cyano-3-phenyl-acrylic acid ethyl ester) as a probe reaction to examine the catalytic performance of the obtained materials. Before the reaction, the diamine-functionalized materials were treated with 0.2 M methanol solution of TMAOH to remove the residue Clions and to neutralize the protonated amine groups in the samples. The analysis showed that the physical properties, texture and amine contents of the catalysts after treatment with TMAOH were almost unchanged. Pure siliceous SBA-15 treated under the same condition showed no catalytic activ-ity.Fig. 8gives the dependency of the yields of a,b-unsat-urated compound on the reaction period over various amine-functionalized SBA-15 materials. No other side products except the a,b-unsaturated compound were detected in the products based on the analysis by GC– MS chromatography. All the catalysts showed rapid reac-tion rates in the ﬁrst 150 min, after this the yields of the product only increased slightly. This might be due to the deactivation of the catalysts by adsorption of formed water and a trace amount of amide product[14]. For the catalysts

with ATMS/(TEOS + ATMS) molar ratio less than 0.15, the product yields increase with the nitrogen content in the catalysts. As the ATMS concentration was further increased to 0.20, the product yield decreased markedly due to the low surface area and pore volume of the mate-rial. It can be seen in Fig. 8 that the product yield over 10ED-P-S is slightly lower than that over SBA-10ED-P. Both contained similar loadings of amino groups, but the latter was prepared without salt and had poor X-ray mesostructure ordering. This might be because the reactants could access the catalytic sites more easily by dif-fusion as the pore channels of SBA-10ED-P catalyst were probably shorter than those of SBA-10ED-P-S.

Since each diamine functional group contains two amine sites but aminopropyl group contains only one amine site, it is expected that aminopropyl-functionalized SBA-10NH2

catalyst should have similar amine sites as diamine-func-tionalized SBA-5ED-P-S catalyst. However, analysis shows that single amine-functionalized SBA-10NH2 has higher

surface area, larger pore volume and higher N content than 5ED-P-S. Nevertheless, the catalytic activity of SBA-10NH2is lower than the diamine-functionalized

SBA-5ED-P-S as shown inFig. 8. The discrepancy is so marked that it is proposed that two amine sites on the diamine group might have synergistic eﬀect to some extent. The detailed mechanism will need further investigation.

The spent SBA-10ED-P-S was treated with methanol in a Soxhlet apparatus for 3 h and then dried at 120C over-night before tested as a catalyst again. The spent catalyst gave only slightly lower activity than the fresh catalyst (85% versus 93% yield), indicating that most of the active sites could be regenerated by a simple solvent treatment. 4. Conclusions

Highly-ordered large pore diamine-functionalized SBA-15 materials with high loadings of amino groups (up to 3.49 mmol g1) were synthesized under strong acid condi-tion by the combinacondi-tion of TEOS prehydrolysis and addi-tion of inorganic salt NaCl. Prehydrolysis of TEOS prior to the addition of ATMS precursors was a key step for the successful synthesis of ordered functionalized materials. Inorganic salt could enhance the mesostructure ordering and inﬂuence the morphology of the materials. The dia-mine-functionalized SBA-15 showed better catalytic per-formance than aminopropyl-functionalized counterpart in the Knoevenagel reaction of benzaldehyde with ethyl cyno-acetate to form a,b-unsaturated compounds in liquid phase.

Acknowledgment

The authors acknowledge the ﬁnancial supports from the National Science Council, Taiwan. Acknowledgements are also extended to Mr. C.-Y. Tang, Ms. C.-Y. Lin and Ms. G.-W. Lu of National Taiwan University for TEM, SEM experiments and elemental analysis.

C O H C CO2Et CN + C H H2C CO2Et CN + H2O Scheme 1. 0 100 200 300 400 500 0 20 40 60 80 100 Yield (%)

Reaction time / min

SBA-5ED-P-S SBA-10ED-P-S SBA-15ED-P-S SBA-20ED-P-S SBA-10ED-P SBA-10NH₂ Spent SBA-10ED-P-S

Fig. 8. Knoevenagel reaction of benzaldehyde with ethyl cyanoacetate in cyclohexane over diﬀerent diamine-functionalized SBA-15 catalysts.

(9)

References

[1] C.T. Kresge, M.E. Leonowicz, W.J. Roth, J.C. Vartuli, J.S. Beck, Nature 359 (1992) 710.

[2] I. Rodriguez, S. Iborra, A. Corma, F. Rey, J.L. Jorda´, Chem. Commun. (1999) 593.

[3] A. Stein, Adv. Mater. (Weinheim. Ger.) 15 (2003) 763.

[4] I.K. Mbaraka, D.R. Radu, V.S.-Y. Lin, B.H. Shanks, J. Catal. 219 (2003) 329.

[5] X.G. Wang, C.C. Chen, S.-Y. Chen, Y. Mou, S. Cheng, Appl. Catal. A: Gen. 281 (2005) 47.

[6] S. Dai, M.C. Burleigh, Y. Shin, C.C. Morrow, C.E. Barnes, Z. Xue, Angew. Chem., Int. Ed. 38 (1999) 1235.

[7] H. Hata, S. Saeki, T. Kimura, Y. Sugahara, K. Kuroda, Chem. Mater. 11 (1999) 1110.

[8] V.S.Y. Lin, C.Y. Lai, J. Huang, S.A. Song, S. Xu, J. Am. Chem. Soc. 123 (2001) 11510.

[9] W. Zhou, J.M. Thomas, D.S. Shephard, B.F.G. Johnson, D. Ozkaya, T. Maschmeyer, R.G. Bell, Q. Ge, Science 280 (1998) 705.

[10] W.H. Zhang, X.B. Lu, J.H. Xiu, Z.L. Hua, L.X. Zhang, M. Robertson, J.L. Shi, D.S. Yan, J.D. Holmes, Adv. Funct. Mater. 14 (2004) 544.

[11] D. Margollese, J.A. Melero, S.C. Christianesn, B.F. Chmelka, G.D. Stucky, Chem. Mater. 12 (2000) 2448.

[12] K.A. Koyano, T. Tatsumi, Y. Tanaka, S. Nakata, J. Phys. Chem. B 101 (1997) 9436.

[13] M.H. Lim, A. Stein, Chem. Mater. 11 (1999) 3285.

[14] D.J. Macquarrie, D.B. Jackson, Chem. Commun. (1997) 1781. [15] M.T. Drexler, M.D. Amiridis, J. Catal. 214 (2003) 136.

[16] G. Sartori, F. Bigi, R. Maggi, R. Sartorio, D.J. Macquarrie, M. Lenarda, L. Storaro, S. Coluccia, G. Martra, J. Catal. 222 (2004) 410. [17] T. Yokoi, H. Yoshitake, T. Tatsumi, J. Mater. Chem. 14 (2004) 951. [18] S. Huh, J.W. Wiench, J.C. Yoo, M. Pruski, V.S.Y. Lin, Chem. Mater.

15 (2003) 4274.

[19] D.J. Macquarrie, D.B. Jackson, J.E.G. Mdoe, J.H. Clark, New J. Chem. 23 (1999) 539.

[20] A.S.M. Chong, X.S. Zhao, J. Phys. Chem. B 107 (2003) 12650. [21] H.H.P. Yiu, P.A. Wright, N.P. Botting, J. Mol. Catal. B: Enzym. 15

(2001) 81.

[22] Y.J. Han, G.D. Stucky, A. Butler, J. Am. Chem. Soc. 121 (1999) 9897.

[23] X.G. Wang, K.S.K. Lin, J.C.C. Chan, S. Cheng, Chem. Commun. (2004) 2762.

[24] X.G. Wang, K.S.K. Lin, J.C.C. Chan, S. Cheng, J. Phys. Chem. B 109 (2005) 1763.

[25] X. Wang, Y.-H. Tseng, J.C.C. Chan, S. Cheng, Micropor. Mesopor. Mater. 85 (2005) 241.

[26] J.M. Kim, S.K. Kim, R. Ryoo, J. Phys. Chem. B 103 (1999) 6200. [27] C. Yu, B. Tian, J. Fan, G.D. Stucky, D. Zhao, J. Am. Chem. Soc. 124

(2002) 4556.

[28] B.L. Newalkar, S. Komarneni, Chem. Mater. 13 (2001) 4573. [29] Y. Wang, B. Zibrowius, C. Yang, B. Spliethoﬀ, F. Schu¨th, Chem.

Commun. (2004) 46.

[30] W.H. Zhang, J.L. Shi, H.R. Chen, Z.L. Hua, D.S. Yan, Chem. Mater. 13 (2001) 648.

[31] C. Yang, P.H. Liu, Y.F. Ho, C.Y. Chiu, K.-J. Chao, Chem. Mater. 15 (2003) 275.

[32] J.L. Gu, J.L. Shi, G.J. You, L.M. Xiong, S.X. Qian, Z.L. Hua, H.R. Chen, Adv. Mater. 17 (2005) 557.

[33] H. Zhu, B. Lee, S. Dai, S.H. Overbury, Langmuir 19 (2003) 3974. [34] D. Jiang, Q. Yang, J. Yang, L. Zhang, G. Zhu, W. Su, C. Li, Chem.

Mater. 17 (2005) 6154.

[35] J.A. Melro, G.D. Stucky, R. van Grieken, G. Morales, J. Mater. Chem. 12 (2002) 1664.

[36] Y.Q. Wang, C.M. Yang, B. Zibrowius, B. Spliethoﬀ, M. Linden, F. Schuth, Chem. Mater. 15 (2003) 5029.

[37] Y. Mori, T.J. Pinnavaia, Chem. Mater. 13 (2001) 2173.

[38] D. Zhao, J. Feng, Q. Huo, N. Melosh, G.H. Frodrickson, B.F. Chmelka, G.D. Stucky, Science 279 (1998) 548.

[39] X.S. Zhao, G.Q. Lu, A.K. Whittaker, G.J. Millar, H.Y. Zhu, J. Phys. Chem. B 101 (1997) 6525.