Available online at www.sciencedirect.

com

Microporous and Mesoporous Materials 115 (2008) 469479 www.elsevier.com/locate/micromeso

Calcination and thermal degradation mechanisms of triblock copolymer template in SBA-15 materials
F. Berube, S. Kaliaguine *
Chemical Engineering Department, Laval University, Que., Canada G1V 0A6 Received 5 December 2007; received in revised form 8 February 2008; accepted 19 February 2008 Available online 4 March 2008

Abstract The calcination under air and degradation under inert atmosphere of as made SBA-15 surfactant templated mesostructured silica materials were studied using a combination of N2 sorption at 196 C, mass spectrometry (MS) monitored temperature programmed oxidation and degradation, thermogravimetric analysis (TGA), 13C MAS NMR and Fourier transform infrared (FTIR) spectroscopy. The characterization of the materials treated at dierent temperatures under oxidative and inert atmospheres indicated that both processes follow stepwise mechanisms. SBA-15 materials exhibit three families of pores: primary main mesopores, complementary intrawall mesopores (>2 nm) and intrawall micropores (<2 nm). Under oxidative atmosphere, the primary mesopores and the larger framework intrawall pores are rst emptied below 200 C with the production of volatile organic compounds (VOCs). This step is followed by an oxidation of the PEO chains from the intrawall micropores (<2 nm) producing CO2 by combustion. Under inert atmosphere, the degradation of the organic template also begins rst in the primary mesopores. However, an increase in the pore diameter up to 550 C indicates that the complete liberation of primary mesopores is much slower than for calcination under air and occurs simultaneously with the removal of the PEO chains occluded within framework micropores. 2008 Elsevier Inc. All rights reserved.
Keywords: SBA-15; Triblock copolymer; Calcination; Thermal degradation; Template removal

1. Introduction Since the discovery of MCM-type mesostructured materials [1], many applications of these materials have been proposed in elds as diverse as catalysis [2,3], separation [4], electronic [5], optoelectronics [6] and nanocasting [7 9]. By using the surfactant templating pathways, substantial eorts have been deployed to synthesize new materials with dierent mesostructures and pore sizes [1013]. In 1998, Zhao et al. developed a new mesostructured material by using non ionic triblock copolymers as structuring agent in acidic medium [14,15]. Compared with the previous M41S family, the SBA-15 type silica shows an enhanced hydrothermal stability because of thicker and more condensed silica walls [16,17].
*

Corresponding author. Tel.: +1 418 656 2708; fax: +1 418 656 3810. E-mail address: serge.kaliaguine@gch.ulaval.ca (S. Kaliaguine).

Nevertheless, several aspects of the synthesis of these materials remain to be improved to develop more valuable solids for a wider range of applications. An important aspect of the synthesis of these mesostructured materials is the removal of the organic template occluded within the pores. Several methods have been developed to remove the organic template from hybrid mesophase such as oxidative ozone treatment [18,19], supercritical uid extraction [20,21], microwave digestion [22] and ether cleavage by an acid [23,24], but the most popular ones are calcination under air and extraction with an organic solvent [2529]. Several studies have been carried out to understand the inuence of calcination on the physico-chemical properties of mesostructured SBA-15 type silica materials and showed that signicant lattice shrinkage occurs upon high-temperature treatment [20,22,23,25,30,31]. However, relatively little works was done to understand the mechanism of the organic template oxidation along the calcination process [15,30].

1387-1811/$ - see front matter 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.micromeso.2008.02.028

470

F. Berube, S. Kaliaguine / Microporous and Mesoporous Materials 115 (2008) 469479

Several studies demonstrated that SBA-15 materials may contain intrawall pores that interconnect the ordered primary mesopores [3235]. This porous framework is formed as the hydrophilic part of the triblock copolymer is occluded within the silica walls. These interactions between the inorganic framework and the PEO chains appear to strongly aect the calcination behavior. Kleitz et al. showed by using TG-DTA/MS that the removal of the triblock copolymer under air is a stepwise mechanism [30]. These authors showed that a decomposition step between 150 and 200 C is responsible for the transformation of the major part of the polyalkene oxide into carbonaceous species. From XRD measurements at varying calcination temperature, they also showed that this rst step was accompanied by a strong increase in the intensities of the low angle reections. From 200 to 250 C, the oxidation of the block copolymer produced CO2 and H2O. Interestingly, a noticeable decrease in the (1 0 0) reection intensity was recorded over this temperature range and was attributed to a decrease in the scattering contrast due to the removal of the organic compounds from the intrawall pores. Moreover, by considering the relatively low temperature decomposition of the triblock copolymer (e.g. 150180 C) compared to that of cationic surfactants, it was also suggested that oxidation of the block copolymer is catalyzed by the inorganic framework [15,30]. The removal of the organic structure directing species by a thermal treatment under inert atmosphere is also commonly used in laboratories. The popularity of this method has recently increased with the discovery of the mesoporous polymers/carbons synthesized by a direct surfactant templating pathway [36,37]. In this case, the thermal degradation under inert atmosphere could be a good alternative to remove the structuring agent because this new class of mesostructured materials are highly sensitive to an oxidative atmosphere. The present work is aiming at understanding the mechanisms of template removal either by calcination under oxygen or thermal degradation under inert atmosphere in the SBA-15 materials. The template removal was studied by a combination of N2 sorption at 196 C, TGA, MS monitored temperature programmed oxidation and degradation, 13C MAS NMR and FTIR. 2. Experimental 2.1. Materials SBA-15 material was obtained following the procedure reported by Zhao et al. [14,15]. Pluronic P123 (BASF, Mw = 5800 g/mol) was used as the template and tetraethyl orthosilicate (Aldrich) as the silica source. In a typical synthesis, 7.659 g of P123 was dissolved in 290 mL of a 1.6 M aqueous HCl solution. 16 g of TEOS was then added dropwise. The synthesis was carried out for 20 h at 35 C followed by 24 h of hydrothermal treatment at 80 C. The solid products were recovered by ltration and dried under

air at 80 C during 24 h. Prior to the calcination and degradation experiments, all the samples were again dried at 80 C under vacuum overnight to remove the physisorbed solvent or water. 2.2. Characterization X-ray diraction (XRD) patterns were recorded on a Bruker D4 X-ray diractometer with Ni-ltered Cu Ka radiation (40 kV, 40 mA). Elemental analyses of carbon were performed with a Carloerba instrument 1500 Carbon, Nitrogen, Sulfur analyser. Nitrogen adsorption/desorption isotherms at 196 C were determined using a QUANTACHROME NOVA 2000 instrument. Before the experiments, the made sample was evacuated at 120 C for 6 h and thermally treated samples were evacuated at 150 C for 4 h. The specic surface area was calculated using the BET equation over the relative pressure range of 0.10.2. The micropore and mesopore volumes were estimated from the as-plot using a LiChrospher Si-1000 (EM Separations, Gibbstown, NJ) as reference material [38]. The pore size distributions were determined from the adsorption branch of isotherms using the modied BJH method [39]. Temperature programmed calcination and temperature programmed degradation monitored by mass spectrometry (TPC-MS and TPD-MS) were performed using a RXM100 multi catalyst testing and characterization system (Advanced Scientic Design Inc.). Seventy milligram of support was placed in a U-shaped reactor coupled with a quadrupole mass spectrometer. All the experiments were carried out under a ow (50 mL/min) of 10% O2 in He (TPC) or 100% He (TPD) from 25 C to the various temperatures of interest with a 2 C/min ramp and then cooled down to room temperature under the same ow. The temperature was maintained for dierent times after the end of the ramp to achieve polymer degradation and oxidation. The isothermal periods used during calcination and degradation of the dierent samples are presented in Table 1. Masses m/z 12, 14, 16, 18, 20, 26, 2730, 3646 and 58 were recorded using a mass spectrometer (UTI) to identify the
Table 1 Experimental conditions used during calcination (C) and thermal degradation (D) of as-synthesized SBA-15 Samples (C,D-TC) C160 C175 C270 C335 C575 D360 D450 D550 D700 Ramp (C/min) 2 2 2 2 2 3 3 3 3 Isothermal time (min) 0 120 120 120 15 120 120 120 20 Flow composition 10% 10% 10% 10% 10% He He He He O2 O2 O2 O2 O2 in in in in in He He He He He

 F. Berube, S. Kaliaguine / Microporous and Mesoporous Materials 115 (2008) 469479

471

dierent combustion products (VOCs, CO, CO2 and H2O). Water and higher molecular weight hydrocarbon species were trapped at 80 C (dry ice in ethanol solution) before reaching the mass spectrometer leak in order to increase the monitoring resolution. Thermogravimetric analysis (TGA) was carried out under owing air or nitrogen on a Q-500 TA instrument at a heating rate of 2 C min1. 13 C NMR spectra were obtained at room temperature on a BRUKER AVANCE 300 MHz spectrometer. The experimental conditions were a recycle delay of 3 s, a number of scans between 14,000 and 32,000, contact time of 1 ms and 1H p/2 pulse of 3.9 ls. Infrared spectra were recorded using a BIO-RAD FTS60 FTIR spectrometer with 6 mg wafer except for Pluronic P123 which was slightly spread on a 100 mg KBr wafer. 3. Results and discussion 3.1. Calcination under air Fig. 1 shows XRD pattern of SBA-15 material calcined at 575 C. The diractogram shows three diraction peaks associated with the (1 0 0), (1 1 0) and (2 0 0) reections consistent with the two-dimensional (2-D) hexagonal structure. The preparation conditions of SBA-15 samples calcined and thermally degraded at various temperatures are presented in Table 1. Some physico-chemical properties of the calcined materials are reported in Table 2. Compared to the as made material which presents a carbon content of approximately 30%, the SBA-15 calcined at 160 C without isothermal period shows a carbon loss of 18% indicating that under air, the oxidation of P123 begins at relatively low temperature. By calcining the material at 175 C, more than 85% of the initial organic template was removed and for materials calcined at higher temperatures, the residual carbon contents gradually decrease indicating that the complete elimination of organic template requires high temperature treatment (T > 500 C).

Table 2 Structural properties of SBA-15 calcined at various temperatures Samples As-synthesized C160 C175 C270 C335 C575
a b

%Ca 29.7 24.2 4.3 1.4 0.8 0.3

SBET (m2/g) 40 170 1020 1120 1080 900

Dpb (nm) 7.7 8.3 8.8 9.0 8.5 8.3

Vtc (cc/g) 0.09 0.31 1.12 1.19 1.16 1.03

Vmesd,e (cc/g) 0.05 0.26 0.79 0.84 0.81 0.75

Vmicd (cc/g) 0 0 0.14 0.17 0.17 0.16

Determined by elemental analysis. Pore diameter determined by the modied BJH method from the adsorption branch. c Adsorbed volume at P/Po = 0.995. d Micropore and mesopore volumes determined from as plot [38]. e Vmes+mic (sum of mesopore and micropore volumes, 1.6 < as < 2.0) Vmic (micropore volume, 0.9 < as < 1.2).

Figs. 2 and 3 show respectively N2 sorption isotherms at 196 C and an evolution of the various pore volumes of a SBA-15 sample calcined at dierent temperatures. The as made material shows a very weak adsorption capacity indicating that almost all the porous network is lled by the organic template. After calcination at 160 C, alpha-s plot shows that no microporous volume is free although the mesopore volume is 31% of its maximal value obtained for sample calcined at 270 C. This result suggests that the organic template located inside the primary mesopores is oxidized before the PEO chains which are associated with intrawall microporosity. By an X-ray diraction analysis of SBA-15 materials at varying calcination temperature, Kleitz and coworkers noted an increase in the intensity of the reexions (1 0 0), (1 1 0) and (2 0 0) of the diffractogram up to 200 C followed by a short decrease between 200 and 250 C [30]. This uctuation of the diraction peaks intensities was attributed to the successive releases of the mesopores and intrawall porosity and is thus conrmed by our N2 sorption results. After a 2 h isotherm

1800 1600 1400

(a) (b) (c) (d) (e) (f)

Vads (cc/g)

1200 1000 800 600 400 200 0 0.0

Signal (A.U.)

0.2

0.4

0.6

0.8

1.0

P/P0
1 2 3 4 5
Fig. 2. N2 sorption isotherm at 196 C obtained on as-synthesized SBA15 (a), C160 (b), C175 (c), C270 (d), C335 (e), C575 (f). The rst ve isotherms are shifted, respectively by 1600, 1400, 770, 470 et 200 cc/g.

2 theta
Fig. 1. XRD pattern of SBA-15 calcined at 575 C.

472

F. Berube, S. Kaliaguine / Microporous and Mesoporous Materials 115 (2008) 469479

1.0

1.0
0.8 0.6 0.4 0.2 0.0 0 100 200 300 400 500 600

V/2dr [cc/(nmg)]
Calcination temperature (C)

0.8 0.6 0.4 0.2 0.0 2 4 6 8 10

V (cc/g)

Fig. 3. Comparative plots of micropore volumes (solid circles) and mesopore volumes (open circles) of SBA-15 calcined at dierent temperatures.

Dp (nm)
Fig. 4. Pore size distributions (PSDs) calculated from the nitrogen adsorption isotherms by modied BJH method on SBA-15 calcined at dierent temperatures.

at 175 C, 85% of the micropore volume and 95% of the mesopore volume are free, compared to their maximal values. Although the removal of the copolymer begins in the primary mesopores, a very high micropore volume for this sample indicates that the oxidation reactions evolve rapidly in the two types of pore systems. Between 175 and 270 C, an increase of the pore volume made up to 64% of total mesopores and 36% of total micropores is also observed. Then, for higher temperatures (>270 C), mesopore and total pore volumes decrease suggesting that the lattice shrinkage occurs for these temperatures in accordance with the results of Kleitz et al. [30]. Surprisingly, the micropore volume remains approximately unchanged for temperature higher than 270 C even if, according to Ref. [30], lattice shrinkage occurs in these conditions. It may be conjectured that since organic residues are still present in these micropores at 335 C and even 575 C (see Table 2), the almost unchanged micropore volume could result from the emptying of some micropores occurring simultaneously with pore shrinkage. Pore size distributions (PSDs) of the same series of calcined samples are shown in Fig. 4. Prior to calcination, the sample shows only a small broad distribution around 7.7 nm. In addition to a broad distribution around 8.1 nm, the PSD of sample calcined at 160 C shows another type of porosity around 2 nm. Previous studies showed that an hydrothermal treatment could transform intrawall microporosity into a porosity of higher size interconnecting the primary mesopores [32,33]. It was also proven that an aging temperature of 80 C leads to an appreciable secondary mesopore volume within the silica walls [4,33]. The sample calcined at 175 C presents a narrow distribution around 9 nm which is not signicantly shifted for material calcined at 270 C indicating that the oxidation reactions in the primary mesopores are almost nished for the sample calcined at lower temperature. However, the comparison between the PSDs of these two samples (C175 and C270) reveals a slight increase in the

volume of pores smaller than 4 nm for the latter sample. This result indicates that between these temperatures, the removal of the organic template is located exclusively within the silica walls. Finally, one can conclude that the lattice contraction begins around 300 C with a decrease in the pore diameter to 8.3 nm for sample calcined at 335 C. The TPC under 10% O2 in He and TGA under air results are presented in Figs. 5 and 6, respectively. The analysis of oxidation products by mass spectrometry shows that the template removal under oxidative atmosphere is a stepwise mechanism. An important peak rst appears between 125 and 190 C. One also notices that this step is associated with the main weight loss of 37%, as reported before [30]. Mass spectra of the combustion products at 130 C also showed that masses m/z 14, 26, 27, 29, 30, 43 and 58 are present in addition to 12, 28 and 44 associated with CO and CO2. As reported before by Kleitz and coworkers, this rst calcination step is associated with a fragmentation of the block copolymer in volatile organic compounds (VOCs) [30]. Mass m/z 29 is the most important VOC fragment indicating that the carbonaceous species formed during that step may have a carbonyl group. Moreover, a study made by Decker and Marchal showed that the degradation of diethylene glycol in the presence of oxygen produced aldehydes, carboxylic acids and alcohols by a mechanism involving radical intermediates [40]. The comparison between the TGA of as made material and that of bulk pluronic P123 copolymer reveals that the silica mesostructure seems to have a signicant eect on the rate of this polymer fragmentation process (see Fig. 6). This result could be explained by a better dispersion of the organic phase in the material making it more accessible to oxygen. Furthermore, the low thermal conductivity of the silica walls may induce a temporary rise in local temperature. After this rst polymer fragmentation step, TPC monitored by m/z 28 and 44 also shows a shoul-

 F. Berube, S. Kaliaguine / Microporous and Mesoporous Materials 115 (2008) 469479

473

m/z 12 m/z 28 m/z 44

m/z 14 m/z 26 m/z 29

100

200

300

400

500

600

Temperature (C)
Fig. 5. MS monitored temperature programmed oxidation of as-synthesized SBA-15. The bottom graph scale is reduced by a factor of two compared to the upper one.

140 120
SBA-15

4 3 2 1
P123

100 80 60 40 20 0 0 100 200 300 400

0 500

Temperature (C)
Fig. 6. TGA and DTG curves under air of as-synthesized SBA-15 and Pluronic P123. The upper TGA and DTG curves are shifted, respectively by 50% and 3.1%/C.

der between 190 and 300 C probably due to the oxidation of the carbonaceous species produced at lower temperature. Finally, TPC followed with m/z 44 clearly reveals another peak between 300 and 500 C which may be associated to the oxidation of remaining organic template in strong interaction with the silica walls. Fig. 7 shows 13C NMR spectra of the same SBA-15 materials following their respective thermal post-treatment. As reported before, chemical shifts of 19.3, 73 and 75 ppm are associated with polyoxopropylene and the shoulder at

72 ppm is attributed to the hydrophilic chains of the copolymer template [23,24]. The sample calcined at 160 C clearly shows a decrease of PPO signal in comparison to PEO lines proving that an ether cleavage of the P123 occurs during the rst calcination step and rst empties the primary mesopores. One also notices the appearance of a peak around 163 ppm which becomes more important for material calcined at 175 C. This line is associated with the carbon of the formic acid terminal groups (O@CHAOA) which are present in polyoxides oxidation products and are characterized by the mass fragment m/z 29 identied by TPC-MS experiments (see discussion above). 13C NMR spectrum of the material calcined at 175 C does not present the P123 characteristic lines (73 and 75 ppm) any more. First, this spectrum shows that essentially no more PPO fragments remain in the material. Then, important lines at 70, 66 and 62 ppm were also obtained for this sample. This triplet, which seems unchanged in the spectrum of material calcined at 270 C, could correspond to carbons associated with ethylene glycol and/or with those located in polymer chains end. The remaining organic species present in the material are composed of small fragments corresponding to the copolymer hydrophilic part and indicate that the oxidation reactions take place in the silica framework at these temperatures. A small peak at 50 ppm is also featured in the spectrum of material calcined at 175 C and becomes more important in that of the material calcined at 270 C. This line corresponds to the methyl group (OACH3) indicating that the polyether oxidation can produce this termination group. Finally, no signals for sample calcined at 335 C were observed. Infrared spectra of the SBA-15 calcined at various temperatures are shown in Fig. 8. All the IR spectra of as made material and those calcined at temperatures lower than 335 C show the presence of pluronic P123 copolymer (13001500 cm1 and 28003050 cm1) with various contents. Moreover, one notices a peak around 1740 cm1 appearing on all calcined materials spectra which corresponds to C@O groups. This line is ascribed to the formic acid termination as previously identied by 13C MAS NMR. Finally, another signal around 1640 cm1 is associated with water physisorbed on the material surface due to air exposure during the sample preparation. The latter was present in all spectra, but was weaker for the as made sample and the sample calcined at 160 C. The presence of triblock copolymer covering the silica surface is responsible for the lower uptake of physisorbed water. Furthermore, the important increase in physisorbed water for samples calcined at higher temperatures is due to some increase in their hydrophilicity likely associated with micropore surface becoming accessible in these materials. 3.2. Thermal degradation under inert atmosphere and comparison with calcination under air The physicochemical properties of SBA-15 materials following their thermal degradation under He at various

Weight loss (% )

Intensity (A.U.)

dm/dT (wt%/C)

474

F. Berube, S. Kaliaguine / Microporous and Mesoporous Materials 115 (2008) 469479

74.3 ppm 72.0 ppm 69.1 ppm 66.2 ppm 62.3 ppm 49 ppm

19.3 ppm

(a) (b)
Signal (A.U.)

163 ppm

(c)

(d) (e)
[CH2CHCH3O]m HOCH2CH2OH O=CHOCH2CH2O [CH2CH2O]nCH2CH2OH OCH3 [CH2CHCH3O]m

200
13

150

100

50

ppm
Fig. 7. C MAS NMR spectra of as-synthesized SBA-15 (a), C160 (b), C175 (c), C270 (d) and C335 (e).

(a)

(b)

Transmittance

(c) (d) (e) (f) (g)

3000 2500 2000
-1

1740 cm-1
1500

wavenumber (cm )
Fig. 8. IR spectra of as-synthesized SBA-15 (b), C160 (c), C175 (d), C270 (e), C335 (f) and C575 (g). Reference IR spectra of P123 spread on a KBr wafer is also shown (a).

temperatures are presented in Table 3. After degradation at 360 and 450 C, 69% and 85% of the organic phase were respectively removed from the materials. The comparison with the calcination under air (see Table 2) reveals that the rate of thermal degradation is lower and it occurs at

higher temperature. Indeed, the materials calcined under air at 175 C or treated under inert atmosphere at 450 C show essentially the same residual organic content, whereas both processes begin at clearly dierent temperatures, namely 125 C and 300 C, respectively. The samples

 F. Berube, S. Kaliaguine / Microporous and Mesoporous Materials 115 (2008) 469479 Table 3 Structural properties of SBA-15 treated under inert atmosphere at dierent temperatures Samples As-synthesized D360 D450 D550 D700
a b

475

1.0 0.8

%Ca 29.7 9.3 4.7 1.3 0.2

SBET (m2/g) 40 710 940 960 320

Dpb (nm) 7.7 8.4 8.4 8.7 6.5

Vtc (cc/g) 0.09 0.83 1.02 1.00 0.39

Vmesd,e (cc/g) 0.05 0.66 0.78 0.74 0.35

Vmicd (cc/g) 0 0.05 0.11 0.15 0

V (cc/g)

0.6 0.4 0.2 0.0 0 200 400 600 800

Determined by elemental analysis. Pore diameter determined by the modied BJH method from the adsorption branch. c Adsorbed volume at P/Po = 0.995. d Micropore and mesopore volumes determined from as plot [38]. e Vmes+mic (sum of mesopore and micropore volumes, 1.6 < as <2.0) Vmic (micropore volume, 0.9 < as < 1.2).

Temperature (C)
Fig. 10. Comparative plots of micropore volumes (solid circles) and mesopore volumes (open circles) of SBA-15 treated under inert atmosphere at dierent temperatures.

obtained after treatments under He at 550 and 700 C have respective carbon contents of 1.3% and 0.2% indicating that the organic template can be removed almost completely under inert atmosphere. Thus, this type of copolymer does not lead to a signicant formation of coke. N2 sorption isotherms and evolution of the various pore volumes of SBA-15 following their degradation at dierent temperatures are presented in Figs. 9 and 10, respectively. 30% of the micropore volume and 85% of the mesopore volume were accessible to N2 for the material treated under He at 360 C in comparison with the maximum values obtained at 550 and 450 C, respectively (see Table 3). Then, the sample treated at 450 C has the highest mesopore volume and 70% of the maximal micropore volume was free of organic template. The highest micropore volume (Vmic = 0.15 cc/g) was obtained for the material treated at 550 C, a temperature at which this material undergoes a decrease of 5% in the mesoporous volume due to lattice shrinkage. It seems that the decrease in pore volume is compensated by the template removal. These

1400 1200 1000

(a)

Vads (cc/g)

(b) (c) (d) (e)

0.0 0.2 0.4 0.6 0.8 1.0

800 600 400 200 0

P/P0
Fig. 9. N2 sorption isotherm at 196 C obtained on as-synthesized SBA15 (a), D360 (b), D450 (c), D550 (d), D700 (e). The rst three isotherms are shifted, respectively by 1300, 700 and 300 cc/g.

results showed that the degradation of the organic template begins in the primary mesopores as observed for calcination under air. However, the complete liberation of primary mesopores was much slower under inert atmosphere. Indeed, the whole mesopore volume is free of organic template for sample calcined under air at 270 C (see Fig. 3), while under an inert atmosphere, this release takes place gradually on a broader range of temperature, between 300 and 550 C. Surprisingly, the material treated under He at 550 C and that calcined under 10% O2 in He at 575 C have essentially the same micropore and mesopore volumes (see Tables 2 and 3). One can thus conclude that the removal of the organic template from SBA-15 material under air and inert atmosphere lead to materials exhibiting the same structural and textural properties. Finally, the SBA-15 material treated at higher temperature (D700) shows no microporosity and a decrease of 50% of its maximal mesopore volume. Zhang et al. investigated the eect of high temperature treatments on physico-chemical properties of SBA-15 materials and showed that a treatment under inert atmosphere at high temperature (900 C) leads to a complete elimination of the material microporosity and causes dramatic shrinkage of the mesopores [41]. The authors explained that hydrolysis and dehydrolysis reactions can lead to the recombination of SiAOASi on the amorphous walls especially on the micropore surface, where silanol density is higher. Pore size distributions (PSDs) of materials treated at various temperatures under inert atmosphere are presented in Fig. 11. The material treated at 360 C presents a narrow distribution centered at 8.1 nm. Moreover, the distribution between 2 and 7 nm reveals that this material contains an appreciable secondary mesopore volume. PSD of SBA-15 material treated at 450 C also exhibits a peak around 8.1 nm which is characterized by a slight widening towards higher pore sizes. At 550 C, the primary mesopore distribution is not shifted to lower value in spite of the important lattice shrinkage occurring at this temperature. The

476

F. Berube, S. Kaliaguine / Microporous and Mesoporous Materials 115 (2008) 469479

600
1.0 0.8 0.6
As-made D360 D425 D550 D700

m/z 29

m/z 15 m/z 29 m/z 44 m/z 58

400
m/z 15

V/2dr [cc/(nmg)]

200 0 250

m/z 30

m/z 58

A
m/z 26 m/z 27 m/z 28

0.4

m/z 28

Intensity (A.U.)

200 150 100

m/z 26 m/z 27

0.2 0.0 2 4 6 8 10

Dp (nm)
Fig. 11. Pore size distributions (PSDs) calculated from the nitrogen adsorption isotherms by modied BJH method on SBA-15 treated under He at dierent temperatures.

50 0 80 60 40
m/z 40

B
m/z 41 m/z 42 m/z 39

volume of intrawall pores between 2 and 4 nm also gradually increases up to this temperature. These results indicate that the template degradation takes place in primary mesopores and larger intrawall pores up to 550 C. Finally, PSD of the material treated at 700 C does not show any more the characteristics of SBA-15 materials. MS monitored temperature programmed degradation under owing He (TPDMS) and TGA under N2 are presented in Fig. 12 and Fig. 13, respectively. A very rich spectrum that contains several mass fragments was observed indicating that the copolymer decomposition produced various volatile organic compounds. The polymer degradation step between 300 and 400 C is associated with the main weight loss determined by TGA and corresponds to the fragmentation of the copolymer into several carbonaceous species [25]. Just like for calcination under air, mass m/z 29 (Fig. 12a) is the most abundant VOCs fragment recorded during that step indicating that the majority of the compounds produced contain a carbonyl group. Then, the mass fragments between m/z 37 and 42 (Fig. 12c) seem to evolve for the same product mainly formed between 300 and 550 C. Due to the chemical structure of the hydrophobic part of the polyalkeneoxide, one can consider propene to be formed during the degradation process (see Supplementary information). Lastly, another compound with m/ z 26, 27, 28 (Fig. 12b) is observed predominantly between 500 and 600 C. These mass fragments are probably due to the presence of ethylene in the gas phase which is expected to be formed from the hydrophilic part of the copolymer (see Supplementary information). These previous results showed that the degradation of the block copolymer leads to successive removal of PEO/PPO chains from the mesostructured silica. Finally, a comparison between TGA under N2 of P123 and as synthesized SBA15 material reveals that the silica mesostructure is responsible for the extension of the polymer degradation at higher

m/z 37 m/z 38 m/z 39 m/z 40 m/z 41 m/z 42

20 0 200

m/z 38 m/z 37

C
400 500 600 700

300

Temperature (C)
Fig. 12. MS monitored temperature programmed degradation (TPDMS) of as-synthesized SBA-15.

temperature. Indeed, TGA of P123 under N2 only featured a single steep weight loss between 300 and 360 C while this weight loss was extended up to 500 C when the polymer is dispersed in mesostructured silica. Fig. 14 shows the 13C MAS NMR spectra of SBA-15 materials treated at various temperatures. In comparison to the as made material, the spectrum of sample treated at 360 C presents a decrease in the relative signal associated with the hydrophobic part of the copolymer (18.5, 73 and 75 ppm) compared to the hydrophilic part (72 ppm). Similarly to the calcination under air process, PPO chains are rst involved during the degradation under inert atmosphere. In addition to the lines corresponding to P123, several other peaks appear on the 13C MAS NMR spectrum of the material treated at 360 C. A line around 50 ppm is associated with methoxy termination group (OACH3) and two other peaks at 60.2 ppm and 16.1 ppm correspond to the carbons of the ethoxy group (OACH2ACH3). Signals appearing at 8.1 and 25.1 ppm are associated with ACH3 of propoxy (OCH2ACH2ACH3) and isopropoxy (OACH(CH3)2) groups, respectively. Finally, a line at 65 ppm could correspond to a carbon having CH2A and AOH groups as neighbors. All the signals associated with the previous chemical species are also identied on the 13C MAS NMR spectrum of material

 F. Berube, S. Kaliaguine / Microporous and Mesoporous Materials 115 (2008) 469479

477

140

4 3 2 1
P123

Weight loss (%)

120 100 80 60 40 20 0 0 100 200 300 400

0 500

dm/dT (wt%/C)

SBA-15

sphere. Indeed, several groups associated with the polymer hydrophobic part (OCH2ACH@CH2, OCH2ACH2ACH3 and OACH(CH3)2) still remain in the material treated under He at 450 C, whereas almost all the PPO chains are removed after the rst calcination step between 150 and 175 C (see Fig. 7). The infrared spectra of materials treated under inert atmosphere at various temperatures are presented in Fig. 15. As for materials calcined under air, the P123

(a) (b)
Transmittance

Temperature (C)
Fig. 13. TGA and DTG curves under N2 of as-synthesized SBA-15 and Pluronic P123. The upper TGA and DTG curves are shifted, respectively by 50% and 3.1%/C.

(c) (d) (e) (f)

1740 cm-1

treated at 450 C, but one notices an increase in their relative intensities compared to the initial P123 signal. In addition to these lines, a peak around 136 ppm corresponds to a OACH2ACH@CH2 group which could explain the production of propene between 300 and 550 C during the degradation process (see discussion above). Finally, no signals appear in the 13C MAS NMR spectrum of the material treated at 550 C. These results reveal a major dierence between the mechanisms of calcination under air and degradation under inert atmo-

3000

2500

2000
-1

1500

Wavenumber (cm )
Fig. 15. IR spectra of as-synthesized SBA-15 (b), D360 (c), D450 (d), D550 (e) and D700 (f). Reference IR spectra of P123 spread on a KBr wafer is also shown (a).

136 ppm

(a)

75.4 ppm 71.4 ppm 25.1 ppm 16.1 ppm 65.6 ppm 8.1 60.2 ppm ppm 49.5 ppm

(b)

(c)

(d)

(PEO)
OCH2CH=CH2 OCH2CH3 OCH(CH3)2 OCH3

[CH2CH2O]nCH2CH2OH OCH2CH2CH3

(PPO)
200 150 100

[CH2CHCH3O]m

50

ppm
Fig. 14.
13

C MAS NMR spectra of as-synthesized SBA-15 (a), D360 (b), D450 (c) and D550 (d).

478

F. Berube, S. Kaliaguine / Microporous and Mesoporous Materials 115 (2008) 469479

signals (13001500 cm1 and 28003050 cm1) decreased as the treatment temperature increased. Moreover, a signal associated with (C@O) groups at 1740 cm1 was also observed, but its relative intensity is much weaker than for samples calcined under air (see Fig. 8). This result can be attributed to the dierence in the species produced during the two processes. IR spectra of samples treated under inert atmosphere show that the relative intensity of the signal corresponding to water physisorbed on the material surface (1640 cm1) also increases as a function of the temperature up to 450 C. Finally, the sample treated at 700 C shows a decreased surface hydrophilicity attributed to the decreased silanol density mostly by condensation of the hydrophilic micropore surface. 4. Conclusions The above results lead to a better comprehension of the combustion and the degradation mechanisms of the triblock copolymer in SBA-15 materials. TGA and the analysis of gas phase products by mass spectrometry allow to identify the various steps of the template removal. Moreover, a combination of N2 sorption at 196 C, 13C MAS NMR and FTIR allowed to monitor the evolution of physico-chemical properties of the as-synthesized SBA15 mesoporous silica with calcination or thermal degradation temperature. All together, these techniques yielded a characterization of the residual template in terms of its chemical composition, quantication and localization in the pore lattice as a function of temperature in the course of both thermal processes. Under an oxidative atmosphere, a fragmentation of the copolymer located in the primary mesopores and the larger framework intrawall pores occurs below 200 C producing several VOCs. Above this temperature, it was found that the residual template is no more associated with the hydrophobic part of the copolymer suggesting the oxidation reactions in larger mesopores have reached completion. N2 sorption at 77K of samples calcined at temperatures higher than 175 C also reveals that the production of CO2 at this temperature is attributed to the oxidation of PEO chains from the intrawall micropores (<2 nm). Under helium, the organic template degradation begins into the primary mesopores around 360 C with the production of several organic compounds. Compared to calcination under air, the production of propene up to 500 C and the chemical groups associated with the PPO chains for sample treated at 450 C both indicate that the removal of template from the primary mesopores is slower and occurs simultaneously with the degradation of the PEO chains from the intrawall pores. The characterizations of the mesoporous materials throughout these processes enable to conclude that the degradation under inert atmosphere and the calcination under air of SBA-15 materials lead to materials having appreciably the same structural prop-

erties. Therefore in situations where traces of residual carbonaceous compounds may be tolerated, it is possible to obtain BET surface area in excess of 1120 m2/g upon calcination at the rather low temperature of 270 C (Table 2). It is indeed possible to calcine the template at higher temperature and produce carbon free SBA-15 but this induces lattice shrinkage with decreased BET surface area and pore diameter. By comparison a maximum in BET surface area and micropore volume is only reached after thermal degradation under helium at 550 C (Table 3). All the above results indicate that the location of the organic template moieties in the mesoporous material has an eect on the reaction products and on the temperature at which these compounds are produced. Thus, by determining the temperature programmed calcination or degradation prole, it becomes possible to specify the localization of the residual partially degraded template by monitoring the sequential pore opening process. As a consequence it might be possible to maintain a complete blockage of the micropores by optimizing the parameters of the calcination process. Acknowledgments We thank NSERC for their nancial support. The authors are grateful to Mr. G. Lemay for assistance in the experimental part. We thank Mr. P. Audet from the Chemistry Department of Laval University for 13C MAS NMR experiments. We thank Prof. F. Kleitz for helpful discussion. Appendix A. Supplementary data Reference mass spectra of ethylene and propene. Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.micromeso.2008. 02.028. References
[1] C.T. Kresge, M.E. Leonowicz, W.J. Roth, J.C. Vartuli, J.S. Beck, Nature 359 (1992) 710. [2] T.-O. Do, D. Desplantier-Giscard, C. Danumah, S. Kaliaguine, Appl. Catal. A Gen. 253 (2003) 543. [3] A. Taguchi, F. Schuth, Micopor. Mesopor. Mater. 77 (2005) 1. [4] V.-T. Hoang, Q. Huang, M. Eic, T.-O. Do, S. Kaliaguine, Langmuir 21 (2005) 2051. [5] D.M. Antonelli, J.Y. Ying, Curr. Opin. Interface Res. 1 (1996) 523. [6] B.J. Scott, G. Wirnsberger, G.D. Stucky, Chem. Mater. 13 (2001) 3140. [7] M. Kruk, M. Jaroniec, R. Ryoo, S.H. Joo, J. Phys. Chem. B 104 (2000) 7960. [8] A.H. Lu, F. Schuth, Adv. Mater. 18 (2006) 1793. [9] H.F. Yang, D. Zhao, J. Mater. Chem. 15 (2005) 1217. [10] Q.S. Huo, D.I. Margolese, U. Ciesla, D.G. Demuth, P.Y. Feng, T.E. Gier, P. Sieger, A. Firouzi, B.F. Chmelka, F. Schuth, G.D. Stucky, Chem. Mater. 6 (1994) 1176. [11] Y. Wan, Y.F. Shi, D. Zhao, Chem. Commun. (2007) 897. [12] Y. Wan, D. Zhao, Chem. Rev. 107 (2007) 2821.

 F. Berube, S. Kaliaguine / Microporous and Mesoporous Materials 115 (2008) 469479 [13] G.J.A.A. Soler-Illia, C. Sanchez, B. Lebeau, J. Patarin, Chem. Rev. 102 (2002) 4093. [14] D. Zhao, J. Feng, Q. Huo, N. Melosh, G.H. Fredrickson, B.F. Chmelka, G.D. Stucky, Science 279 (1998) 548. [15] D. Zhao, Q. Huo, J. Feng, B.F. Chmelka, G.D. Stucky, J. Am. Chem. Soc. 120 (1998) 6024. [16] A.Y. Khodakov, V.L. Zholobenko, R. Bechara, D. Durant, Micropor. Mesopor. Mater. 79 (2005) 29. [17] T. Linssen, K. Cassiers, P. Cool, E.F. Vansant, Adv. Colloid Interface Sci. 103 (2003) 121. [18] G. Buchel, R. Denoyel, P.L. Llewellyn, J. Rouquerol, J. Mater. Chem. 11 (2001) 589. [19] M.T.J. Keene, R. Denoyel, P.L. Llewellyn, Chem. Commun. (1998) 2203. [20] R. van Grieken, G. Calleja, G.D. Stucky, J.A. Melero, R.A. Garcia, J. Iglesias, Langmuir 19 (2003) 3966. [21] S. Kawi, M.W. Lai, Chem. Commun. (1998) 1407. [22] B. Tian, X. Liu, C. Yu, F. Gao, Q. Luo, S. Xie, B. Tu, D. Zhao, Chem. Commun. (2002) 1186. [23] C.-M. Yang, B. Zibrowius, W. Schmidt, F. Schuth, Chem. Mater. 16 (2004) 2918. [24] C.-M. Yang, B. Zibrowius, W. Schmidt, F. Schuth, Chem. Mater. 15 (2003) 3739. [25] M. Kruk, M. Jaroniec, C.H. Ko, R. Ryoo, Chem. Mater. 12 (2000) 1961. [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41]

479

C.-Y. Chen, H.-X. Li, M.E. Davis, Micropor. Mater. 2 (1993) 17. A.G.S. Prado, C. Airoldi, J. Mater. Chem. 12 (2002) 3823. R. Mokaya, W. Jones, J. Mater. Chem. 8 (1998) 2819. S. Hitz, R. Prins, J. Catal. 168 (1997) 194. F. Kleitz, W. Schmidt, F. Schuth, Micropor. Mesopor. Mater. 65 (2003) 1. A. Rumplecker, F. Kleitz, E. Salabas, F. Schuth, Chem. Mater. 19 (2007) 485. R. Ryoo, C.H. Ko, M. Kruk, V. Antochshuk, M. Jaroniec, J. Phys. Chem. B 104 (2000) 11465. A. Galarneau, H. Cambon, F. Di Renzo, R. Ryoo, M. Choi, F. Fajula, New J. Chem. 27 (2003) 73. Z. Liu, O. Terasaki, T. Ohsuna, K. Hiraga, H.J. Shin, R. Ryoo, Chem. Phys. Chem. 4 (2001) 229. M. Imperor-Clerc, P. Davidson, A. Davidson, J. Am. Chem. Soc. 122 (2000) 11925. F. Zhang, Y. Meng, D. Gu, Y. Yan, C. Yu, B. Tu, D. Zhao, J. Am. Chem. Soc. 127 (2005) 13508. S. Tanaka, N. Nishiyama, Y. Egashira, K. Ueyama, Chem. Commun. (2005) 2125. M. Jaroniec, M. Kruk, J.P. Oliver, Langmuir 15 (1999) 5410. M. Kruk, M. Jaroniec, A. Sayari, Langmuir 13 (1997) 6267. C. Decker, J. Marchal, Die Makromolekulare Chemie 166 (1973) 117. F. Zhang, Y. Yan, H. Yang, Y. Meng, C. Yu, B. Tu, D. Zhao, J. Phys. Chem. B 109 (2005) 8723.