Microporous and Mesoporous Materials 147 (2012) 252–258

Contents lists available at ScienceDirect

Microporous and Mesoporous Materials

journal homepage: www.elsevier .com/locate /micromeso

Synthesis of ordered mesoporous silicon oxycarbide monoliths via preceramicpolymer nanocasting

Xiaoyan Yuan a,b, Helin Jin a, Xingbin Yan a,⇑, Laifei Cheng b, Litian Hu a, Qunji Xue a

a State Key Laboratory of Solid Lubrication, Lanzhou Institute of Chemical Physics, Chinese Academy of Sciences, Lanzhou 730000, Chinab National Key Laboratory of Thermostructure Composite Materials, Northwestern Polytechnical University, Xi’an Shaanxi 710072, China

a r t i c l e i n f o a b s t r a c t

Article history:Received 20 April 2011Received in revised form 22 June 2011Accepted 23 June 2011Available online 30 June 2011

Keywords:SiOC ceramicMonolithMesoporousNanocastingPolymer-derived ceramic

1387-1811/$ - see front matter � 2011 Elsevier Inc. Adoi:10.1016/j.micromeso.2011.06.025

⇑ Corresponding author. Tel./fax: +86 931 4968055E-mail address: xbyan@licp.cas.cn (X. Yan).

Highly ordered mesoporous silicon oxycarbide (SiOC) monoliths have been synthesized using liquidpoly(hydridomethylsiloxane) (PHMS) as starting preceramic polymer and mesoporous carbon CMK-3as direct template. Monolithic SiOC-carbon composites were generated via nanocasting of PHMS intoCMK-3, pressing without any additive, cross-linking at 150 �C under humid air and subsequent thermol-ysis at 1000 or 1200 �C under argon atmosphere. The carbon template was finally removed by the ther-mal treatment at 1000 �C in an ammonia atmosphere, as a result of the generation of monolithic SiOCceramics with ordered mesoporous structures. The products were characterized by scanning electronand transmission electron microscopes, X-ray diffraction, Fourier transformation infrared spectrometer,X-ray photoelectron spectroscope and nitrogen absorption–desorption analyzer. The as-prepared SiOCmonoliths exhibited crack-free, ordered 2-dimentional hexagonal p6mm symmetry with high specificsurface areas. With increasing the calcination temperature, the ordered mesoporous structure was stillremained and the specific surface area just had a slight reduction from 616 to 602 m2 g�1. Moreover,the porous SiOC monoliths possessed good compression strengths and anti-oxidation properties.

� 2011 Elsevier Inc. All rights reserved.

1. Introduction

Porous monolithic materials have become very popular due tothe good combination of the compact integral structures and theporous microstructures. Large surface areas and multimodal poros-ities are great advantages in many fields, such as electrochemistry,energy storage, separation, chromatography and catalysis [1–4].For instance, porous monoliths can be used in flow through cata-lytic or separation systems and give better performance and higherpermeability compared with packed columns [4,5]. Among the var-ious porous monolithic materials, silicon carbide (SiC) monolithshave attracted great attention because of their excellent mechani-cal strength, thermal stability and functional semiconductor char-acteristics [6–8]. These characteristics make them hold manypotential applications in high-temperature catalysis [4,9], separa-tion [5,10] and semiconductors [7]. However, the drawback of dis-satisfactory oxidation resistance of SiC materials limits theiraerobic high-temperature environment applications.

It has been reported that, after introducing a third componentinto SiC matrix, such as oxygen, nitrogen, boron or aluminum, as-obtained SiC-based ceramics exhibit much better oxidation resis-tance compared with pure SiC ceramic [11–14]. This reinforcement

ll rights reserved.

.

is fundamentally due to the existence of the complex-covalentbonding configuration in these SiC-based ceramics [11]. Amongthe ternary SiC-based ceramics, silicon oxycarbide (SiOC) ceramicshave been the most promising materials with the consideration oflow cost, simplicity of the set-up and environment safety [11,15–17]. Accordingly, much effort has been paid to the design andsynthesis of SiOC monoliths that have high surface area, tailoredporosity and pore interconnectivity. For example, Ye et al. synthe-sized monolithic silica/resorcinol–formaldehyde aerogels and thenconverted them into porous monolithic SiOC materials [18]. Biaset-to et al. prepared microcellular SiOC foams by means of a commer-cially available preceramic polymer as precursor and poly(methylmethacrylate) microbeads as sacrificial fillers [19]. However, theporous SiOC monoliths, prepared by the aerogel- and foam-tem-plating methods, displayed relatively week mechanical strengths.Vakifahmetoglu et al. prepared monolithic SiOC glass with hierar-chical porosity by a one-pot processing method, but the monolithhad a low specific surface area of 137 m2 g�1 and a weak compres-sion strength of about 1.7 MPa [20]. In addition, the above porousSiOC monoliths were disordered in the mesostructure.

Polymer-derived ceramics is a good road to prepare mesopor-ous ceramic materials [21,22]. PHMS, poly(hydridomethylsilox-ane), as a commercially available polymer, is known as a goodpolymer precursor for preparing SiOC ceramics through crosslink-ing and pyrolysis [23,24]. Because it is liquid at room temperature,

X. Yuan et al. / Microporous and Mesoporous Materials 147 (2012) 252–258 253

PHMS has good ability to fill into nano-scale channels of poroustemplates by casting without the aid of any solvent or heatingprocess.

In this study, we describe a facile synthesis of mesoporous SiOCmonoliths with ordered hexagonal p6mm symmetry using PHMSas the preceramic polymer. In our synthesis process, highly or-dered mesoporous silica SBA-15 was employed as the starting tem-plate to prepare its negative replica carbon CMK-3 template. TheCMK-3 template was subsequently casted with a THF solution ofPHMS using a liquid-phase impregnation. via pressing the gel-likepolymer-carbon composites into tablet-like monoliths, cross-link-ing in humid air, pyrolyzing under argon atmosphere and remov-ing carbon template in ammonia, ordered mesoporous SiOCmonoliths with high specific surface areas were obtained. More-over, the samples possessed high compression strengths and goodanti-oxidation properties, which make them be potential candi-dates for various high-temperature applications.

2. Experimental

2.1. Chemicals

Triblock poly(ethylene oxide)-b-poly(propylene oxide)-b-poly(ethylene oxide) copolymer Pluronic P123 (Mw = 5800, EO20PO70

EO20) and Poly(hydridomethylsiloxane) (PHMS) were purchasedfrom Sigma–Aldrich. Tetraethyl orthosilicate (TEOS, analytical re-agent, AR) was purchased from Tianjin Chemical Reagents Com-pany. Others chemicals were purchased from Shanghai ChemicalCompany. All chemicals were used as received state without anyfurther purification. Deionized water was used in all experiments.High-purity argon and ammonia (99.99%) were used in their as-re-ceived state during the ceramic preparation.

2.2. Synthesis of hard templates

Mesoporous silica SBA-15 template was prepared by hydrother-mal synthesis method according to established procedures [25].That was to using P123 as structure directing agent and TEOS asa precursor. The SBA-15 template exhibited two-dimensionalP6mm hexagonal symmetry with a specific surface area of 527m2 g�1 and an average pore diameter of 8.92 nm. Mesoporous car-bon CMK-3 template was synthesized by the nanocasting methodusing sucrose as a precursor and mesoporous silica SBA-15 as ahard template according to the literature [26]. The CMk-3 templatedisplayed a highly ordered 2D hexagonal mesostructure with aspecific surface area of 1530 m2 g�1 and an average pore diameterof 4.03 nm.

2.3. Preparation of mesoporous SiOC monoliths

The mesoporous SiOC monoliths were synthesized by replicatechnique using PHMS as the ceramic precursor and mesoporousCMK-3 as the hard template. For a typical synthesis, 1.0 g ofCMK-3 was placed in a flask, dried at 150 �C under a vacuum for4 h and cooled down to room temperature (RT), and then addedin a solution of 3.0 g of PHMS and 10 ml of THF. The mixture wasstirred at RT for one day. After that, THF solvent was removed un-der vacuum to obtain gel-like mixture of PHMS and CMK-3. Thegel-like mixture was divided into four parts (1.0 g for each) andeach part was pressed into a cylindrical monolith using a cylindri-cal mould (Ø = 15 mm) at RT under a low pressure of 5 MPa. Allcomposite monoliths were cross-linked at 150 �C for 20 h undera humid air in a muffle furnace. According to the above procedure,the PHMS–CMK-3 composite monoliths were prepared with thesame mass and shape. After that, as-prepared monoliths were

transferred into a horizontal ceramic tube furnace and subjectedto the thermal treatment in an argon atmosphere at 1000 or1200 �C (at heating rate of 1 �C min�1) for 2 h respectively, to gen-erate SiOC-carbon composite monoliths. Finally, the ceramic-carbon composite monoliths were underwent a final thermal treat-ment in an ammonia atmosphere at 1000 �C (a heating rate of2 �C min�1) for 10 h, to remove the carbon template and generateordered mesoporous SiOC monoliths. The monolithic samples pre-pared at different temperatures were denoted as M–SiOC–T, whereM marked as monolith and T marked as temperature.

2.4. Characterization

The morphology of the final products was observed on a fieldemission scanning electron microscope (FE-SEM, JSM 6701F).Transmission electron microscopy (TEM) measurements were con-ducted on a JEM-2010 microscope operated at 200 kV, to reveal theordered structure of the samples. Powdery samples were first dis-persed in ethanol with the aid of sonication and then collectedusing carbon-film-covered copper grids for TEM analyses. Powdersmall-angle X-ray diffraction (SA-XRD) and wide-angle X-ray dif-fraction (WA-XRD) patterns were achieved using a Philipps X’PertPRO X-ray diffraction system (Cu Ka radiation, 0.15406 nm). Thechemical compositions of the samples were analyzed by Fouriertransformation infrared spectroscopy (FTIR) using a Bruker IFS66 VFTIR spectrometer and X-ray photoelectron spectroscopy (XPS)performed on an X-ray photoelectron spectroscope (ESCALAB210VG Scientific). The XPS measurements used Al–Ka radiation(photon energy 1476.6 eV) as the excitation source and the bindingenergy of Au (Au 4f7/2: 84.00 eV) as the reference. Before the XPSscanning, the samples were etching using Ar ions for 30 s. Nitrogenabsorption–desorption isotherm measurements were performedon a Micrometitics ASAP 2020 volumetric absorption analyzer at�196 �C. The Brunauer–Emmett–Teller (BET) method was utilizedto calculate the specific surface area of each sample, and the pore-size distribution was derived from the absorption branch of thecorresponding isotherm using the Barrett–Joyner–Halenda (BJH)method. The total pore volume was estimated from the amountadsorbed at a relative pressure of P/P0 = 0.99. The mechanical com-pressive strength of the mesoporous SiOC monoliths was evaluatedusing a universal tensile testing machine (SHIMADZU UniversalTesting Machine AGS-X 5kN) at RT. The oxidation stability wasevaluated by thermogravimetric analysis (TGA) from RT to1000 �C under air atmosphere with a heating rate of 10 �C min�1.

3. Results and discussion

The synthetic approach for preparing mesoporous SiOC mono-liths is shown in Scheme 1. Herein, the choice of CMK-3 as the hardtemplate is justified by the connection of its hexagonal array ofmesopores by micropores in the walls, which permits the replica-tion process [27]. The main advantage of CMK-3 is that the carbonframework can be removed completely under ammonia at a hightemperature of 1000 �C [13]. Because this NH3 etching process isrelatively slow, the ordered mesoporous structure can be remainedto greatest possible degree [13]. The choice of PHMS as the precur-sor is due to that it has high ceramic yield (>85%), chemical inert-ness under ambient conditions and complete conversion intoceramic at temperature less than 800 �C. In our synthesis, THFsolution of PHMS was easily filled into the nano-scale channelsof CMK-3 template by nanocasting because the CMK-3 has highspecific surface area and large pore volume. After removing THFsolvent slowly, the resulting mixture consisted of PHMS andCMK-3 became a plasticene-like gel, which was easily pressing intomonolith. Furthermore, the presence of Si–H groups in PHMS, hu-

Scheme 1. Synthetic approach for ordered mesoporous SiOC monoliths.

254 X. Yuan et al. / Microporous and Mesoporous Materials 147 (2012) 252–258

mid air atmosphere and sufficient aging time were helpful to thecross-linking of PHMS molecules located within the pores ofCMK-3 and surrounding CMK-3 particles. The formation of Si–OHgroups (shown in Scheme 1) was beneficial to the cross-linkingand the formation of a high SiOC ceramic yield after pyrolysis.Polymers with High ceramic yields are easy to densify and havelow tendency to form cracks when they are applied in constrainedgeometries [23,24]. Thus, through the subsequent polymer-to-ceramic conversion and template-removing process, the connec-tion of hexagonal array of SiOC mesopores and the integrity ofmonolithic SiOC were conserved, as a result of the generation of or-dered mesoporous SiOC monoliths.

Photographs of our typical, as-synthesized M–SiOC-1200showed a good bulk macroscopic appearance without any crackto be seen by the naked eye (Fig. 1a and b). The monolith was sta-ble, even after being immerged into water and sonicated for sev-eral minutes using a low-power ultrasonic bath, there was nocrumbling peeling off from the monolithic body. Because thesizes/or shapes of the monoliths can be easily adjusted by choosingmoulds with different inner diameters/or shapes, the synthetic ap-

Fig. 1. Photographs of the M–SiOC-1200 (a and b) and cross-section

proach can be useful for the large-scale industrial production of or-dered mesoporous SiOC ceramic monoliths. SEM images (Fig. 1cand d) show that the inner structures of the M–SiOC-1000 andthe M–SiOC-1200 were both constructed from interconnectedsmall particles.

Evidence of the maintenance of ordered hexagonal mesoporesfor the M–SiOC-1000 and the M–SiOC-1200 was evidenced byTEM images. As shown in Fig. 2a and b, two samples both exhibitedcharacteristic arrangement of cylindrical channels in large do-mains, which was similar to the structure of original template ofmesoporous SBA-15. In addition, it is clear that the degree of mes-oporous order in the SiOC monoliths decreased with the increase ofthe calcination temperature.

In order to further verify the ordered mesoporous structure, SA-XRD measurements were employed and Fig. 3a shows the corre-sponding SA-XRD patterns of the M–SiOC-1000 and the M–SiOC-1200. It is clear seen that, after the removal of the carbon template,a distinct diffraction peak at 1.10� was observed in the XRD patternof the M–SiOC-1000 (this peak for the M–SiOC-1200 was at 1.02�).This diffraction peak was indexed as the (1 0 0) reflection of the 2-D

SEM images of the M–SiOC-1000 (c) and the M–SiOC-1200 (d).

Fig. 2. TEM images of the M–SiOC-1000 (a) and the M–SiOC-1200 (b) along the [1 1 0] direction, respectively.

1.0 1.5 2.0 2.5 3.0

Inte

nsity

(a.

u.)

2 theta (degree)

M-SiOC-1000

M-SiOC-1200

(100)

a

20 40 60 80

M-SiOC-1200

M-SiOC-1000

Inte

nsity

(a.

u.)

2 theta (degree)

b

Fig. 3. (a) SA-XRD patterns and (b) WA-XRD patterns of the M–SiOC-1000 and the M–SiOC-1200, respectively.

X. Yuan et al. / Microporous and Mesoporous Materials 147 (2012) 252–258 255

hexagonal p6mm symmetry. The d100 value was calculated to be8.02 nm for the M–SiOC-1000 and 8.65 nm for the M–SiOC-1200,respectively. As a consequence, the cell parameter was deducedto be 9.26 nm for the M–SiOC-1000 and 9.66 nm for the M–SiOC-1200. Fig. 3b shows the typical WA-XRD patterns of the M–SiOC-1000 and M–SiOC-1200. The two samples both displayed

500 1000 1500 2000 2500 3000 3500 4000

40

50

60

70

80

90

460cm-1

O-Si-O

M-SiOC-1200

M-SiOC-1000

1100cm-1

Si-O-SiC-O

Tra

nsm

ittan

ce (

%)

Wavenumber (cm-1 )

Fig. 4. FTIR spectra of the M–SiOC-1000 and the M–SiOC-1200.

amorphous SiOC ceramic nature because no obvious characteristicdiffraction peak existed in the patterns. It should be mentionedthat the broad diffraction peak at 21–23� was associated with theSi–O–C glass and the weak peak at 43� in the XRD pattern for theM–SiOC-1200 was correspond to graphite-like carbon [11,28].

To verify the composition of the products, the infrared spectraof the M–SiOC-1000 and the M–SiOC-1200 were measured. Asshown in Fig. 4, an absorption peak around 1080 cm�1 was corre-sponded to a superposition of the stretching vibrations of Si–O andC–O bonds [18]. The presence of C–O bond is owing to the absorp-tion of CO2 on the surface of sample [29]. Another absorption peakat 460 cm�1 was related to O–Si–O bond [18]. Appearance of twoabsorption peaks at 1630 and 3400 cm�1 was attributed to the cer-tain moisture adsorbed in the channels of mesoporous materials.

The M–SiOC-1000 and the M–SiOC-1200 samples were ana-lyzed by XPS as well. It is found that the XPS full-range scanningspectra of the two samples were very similar. Fig. 5 shows theXPS spectrum of the M–SiOC-1200. The strong signals of Si, O,and C in the XPS spectrum indicate that the sample was mainlycomposed of silicon, oxygen and carbon. It should be mentionedthat, the weak nitrogen signal (about 4.2 at.%) might come fromthe formation of surface nitride during the thermal treatment at1000 �C in ammonia atmosphere. The Si/O/C atomic ratio was cal-culated to be 1.0:3.4:1.1 for the M–SiOC-1000 and 1.0:3.3:2.3 forthe M–SiOC-1200, respectively. Thus, the corresponding empiricalformula of Si1.0O3.4C1.1 for the M–SiOC-1000 and the empirical

1000 800 600 400 200 0

0

2

4

6

8

10

12

14

16

18

Si 2s

Inte

nsity

(a.

u.)

Binding energy (ev)

M-SiOC-1200O 1s

N 1s

C 1s

Si 2p

OKLLCKLL

Fig. 5. XPS survey spectrum of the M–SiOC-1200.

256 X. Yuan et al. / Microporous and Mesoporous Materials 147 (2012) 252–258

formula of Si1.0O3.3C2.3 for the M–SiOC-1200 were obtained. It indi-cates that the M–SiOC-1200 contained more carbon comparedwith the M–SiOC-1000. We believe that carbon template would re-act with the Si–O–C phase at the relatively high temperature(1200 �C) and a certain amount of carbon atoms would bond withSiOC ceramic. Thus, the bonded carbon can not be removed by NH3

etching, as a result of relatively high carbon content in theM–SiOC-1200. Additionally, the reaction between carbon templateand SiOC ceramic would also destroy the ordered mesoporous

108 106 104 102 100 98 960.0

0.2

0.4

0.6

0.8

1.0

1.2

Inte

nsity

(a.

u.)

Binding energy (ev)

Si 2p

Si-C

Si-Oa

108 106 104 102 100 98 960.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

Si-O Si-C

Inte

nsity

(a.

u.)

Binding energy (ev)

Si 2pc

Inte

nsity

(a.

u.)

Fig. 6. Si2p and C1s XPS core level spectra of the M–SiOC-100

structure. Thus, as mentioned in TEM explanation, the degree ofmesoporous order for the M–SiOC-1200 was worse than that ofthe M–SiOC-1000.

Fig. 6 shows the Si2p and C1s core level spectra of the M–SiOC-1000 (a and b) and the M–SiOC-1200 (c and d), respectively. Asshown in Fig. 6a and c, the Si2p spectra displayed two peaks at102.7 eV and 101.6 eV for the M–SiOC-1000 and 102.8 eV and101.7 eV for the M–SiOC-1200, respectively. The peaks at high po-sition (102.7 and 102.8 eV) were assigned to the Si–O bond and thepeaks at low position (101.6 and 101.7 eV) were corresponded tothe Si–C bond [29,30]. As shown in Fig. 6b and d, each C1s XPSspectrum indicated the presence of three components: the mainC–Si bonds (at 283.3 eV for the M–SiOC-1000 and at 283.5 eV forthe M–SiOC-1200), the C–O bonds (at 286.9 eV for the M–SiOC-1000 and at 285.9 eV for the M–SiOC-1200), and the C–C bonds(at 284.6 eV for the M–SiOC-1000 and at 284.2 eV for theM–SiOC-1200) [29,30]. The appearance of Si–O and C–O bondswas agreed with the result of FTIR analysis. Also, it is clear thatthe Si–C content in the M–SiOC-1200 was higher than that in theM–SiOC-1000, which was accordant to the corresponding empiri-cal formulas of two samples.

The nitrogen absorption–desorption isotherms measurementswere performed to characterize the porosity of the mesoporousSiOC monoliths. As seen from Fig. 7a, both isotherms exhibited typ-ical type-IV curves with a clear capillary contraction condensingphenomenon for mesoporous materials at a relative pressure (P/P0) of approximately 0.41–0.98. The specific BET surface area(shown in Table 1) was calculated to be 616 m2 g�1 for the M–SiOC-1000 and 602 m2 g�1 for the M–SiOC-1200, respectively.

292 290 288 286 284 282 280

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

Inte

nsity

(a.

u.)

Binding energy (ev)

C 1s

C-C

C-O

C-Sib

290 288 286 284 282 280-0.2

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Binding energy (ev)

d C-Si

C-C

C-O

C1s

0 (a and b) and the M–SiOC-1200 (c and d), respectively.

0.0 0.2 0.4 0.6 0.8 1.050

100

150

200

250

300

350

400

Qua

ntity

Ade

sorb

ed (

cm3 /g

ST

P)

Relative Pressure (P/P0)

M-SiOC-1000 M-SiOC-1200

a

0 5 10 15 80 100

0.0

0.1

0.2

0.3

0.4

0.5

Frac

tiona

l Por

e V

olum

e (c

m3 g-1

nm-1)

Pore Diameter (nm)

M-SiOC-1000 M-SiOC-1200

b

b

Fig. 7. (a) Nitrogen absorption–desorption isotherms and (b) the pore size distributions of the M–SiOC-1000 and the M–SiOC-1200, respectively.

Table 1Textural data and mechanical property of the samples.

Sample Surface area (m2 g�1) Meso-volume (cm3 g�1) Pore size (nm) DV a (%) q b (g cm-3) Compression strength (MPa)

SBA-15 527 1.18 8.92 – – –CMK-3 1530 1.56 4.03 – – –M–SiOC-1000 616 0.62 3.30 33.8 0.556 15.82M–SiOC-1200 602 0.50 2.74 39.6 0.618 24.30

a DV: means volume reduction (%) of the sample after pyrolysis.b q: means density of the resulting sample.

X. Yuan et al. / Microporous and Mesoporous Materials 147 (2012) 252–258 257

Moreover, the BJH mesoporous size distributions (shown in Fig. 7b)for two samples, calculated from the absorption branch, bothshowed a distribution with an apparent average pore size of3.30 nm and 2.74 nm, respectively. It should be mentioned that,the broad range of relative pressures (0.41–0.98) obtained fromthe nitrogen absorption–desorption isotherms and the poorly re-solve diffraction peaks shown in the SA-XRD patterns both indi-cated the poorly hexagonal mesoporous structure. It was normalphenomena for the mesoporous ceramics undergone high synthe-sized temperatures (>1000 �C). Furthermore, we analyzed the poresize of samples by density functional theory (DFT) approach. Thecorresponding curves of the DFT size distributions of the M–SiOC-1000 and M–SiOC-1200 are shown in Fig. 8. As the curvesshow, the pore sizes for the two samples are mainly within

Fig. 8. DFT pore size distributions of the M–SiOC-1000 and the M–SiOC-1200.

2–7 nm. Also, compared with the M–SiOC-1000, the M–SiOC-1200 exhibits a wider pore size distribution, which is in agreementwith the analytic result come from BET model.

Table 1 summarizes the textural properties of the two samples.It is obvious that the M–SiOC-1000 and the M–SiOC-1200 both dis-played high BET surface areas and high total pore volumes. Also, adecrease in the surface area, the pore volume and the average poresize of the monolithic SiOC samples with the increase of the pyro-lysis temperature was observed, which was mainly due to theshrinkage of CMK-3 framework combined with the densificationof SiOC components. The shrinkage and densification were verifiedby the increase of the volume reduction ratio and the increase ofthe density for the M–SiOC-1200 compared with those for theM–SiOC-1000 (shown in Table 1). In addition, as shown in Table

Fig. 9. TGA curves of the M–SiOC-1000 and the M–SiOC-1200.

258 X. Yuan et al. / Microporous and Mesoporous Materials 147 (2012) 252–258

1, the average compressive strength for the M–SiC-1000 and theM–SiC-1200 was 15.82 and 24.30 MPa, respectively. We believethat the higher compressive strength for the M–SiOC-1200 is prob-ably due to its relatively denser matrix obtained at higher pyrolysistemperature.

The oxidation behavior at high-temperature of the M–SiOC-1000 and the M–SiOC-1200 was studied under air by TGA mea-surements, which were carried out under flowing air atmospherefrom RT to 1000 �C with a heating rate of 10 �C min�1. As shownin Fig. 9, the relatively high weight-loss between 500 �C and650 �C was due to the removal of free/and weakly bonded-carboncompounds. The weight loss up to 1000 �C was about 3.7% for theM–SiOC-1000 and 9.8% for the M–SiOC-1200, respectively. It indi-cates that the M–SiOC-1200 contained more carbon compounds,which was accordant to the result come from XPS analysis. Thus,the ordered mesoporous SiOC monoliths can be considered as sta-ble supports to effectively increase the service life of devices thathave to withstand harsh oxidative and thermal environments.

4. Conclusions

In summery, we have demonstrated a simple method to synthe-size ordered mesoporous SiOC monoliths using liquid PHMS as thestarting preceramic polymer and CMK-3 as the hard template. Theshaping process was realized by a simple pressing at RT withoutany agent or heating. The as-synthesized SiOC monoliths showedordered 2D hexagonal p6mm mesostructures with high surfaceareas, high pore volumes and narrow pore size distributions. Also,the monoliths exhibited good mechanical and anti-oxidation prop-erties. The present work provides an efficient strategy to fabricateothers ordered mesoporous ceramic monoliths in large-scaleindustrial production for practical applications, such as chroma-tography and catalysis.

Acknowledgements

The authors acknowledge the support from the Top HundredTalents Program of Chinese Academy of Sciences, the National Ba-sic Research 973 Program of China (2011CB706603) and the Na-tional Nature Science Foundation of China (51005225).

References

[1] G. Singh, D. Moore, R.F. Saraf, Anal. Chem. 81 (2009) 6055–6060.[2] V. Ruiz, C. Blanco, R. Santamaría, J.M. Ramos-Fernández, M. Martínez-

Escandell, A. Sepúlveda-Escribano, F. Rodríguez-Reinoso, Carbon 47 (2009)195–200.

[3] J. Planeta, D. Moravcová, M. Roth, P. Karásek, V. Kahle, J. Chromatogr. A 1217(2010) 5737–5740.

[4] I. Nischang, O. Brüggemann, J. Chromatogr. A 1217 (2010) 5389–5397.[5] A.M. Siouffi, J. Chromatogr. A 1000 (2003) 801–818.[6] N. Orlovskaya, M. Lugovy, F. Ko, S. Yarmolenko, J. Sankar, J. Kuebler, Compos.

Part B 37 (2006) 524–529.[7] R. Riedel, A. Kienzle, W. Dressler, L. Ruwisch, J. Bill, F. Aldinger, Nature 382

(1996) 796–798.[8] A. Los, V. Los, A. Timoshevskii, J. Electro. Mater. 39 (2010) 545–553.[9] G. Winé, J.P. Tessonnier, S. Rigolet, C. Marichal, M.J. Ledoux, C. Pham-Huu, J.

Mol. Catal. A-Chem. 248 (2006) 113–120.[10] C. Sanchez, B. Julian, P. Belleville, M. Popall, J. Mater. Chem. 15 (2005) 3559–

3592.[11] Y.F. Shi, Y. Wan, Y.P. Zhai, R.L. Liu, Y. Meng, B. Tu, D.Y. Zhao, Chem. Mater 19

(2007) 1761–1771.[12] H. Matsuura, K. Aso, S. Kagamihara, H. Iwata, T. Ishida, K. Nishikawa, Appl.

Phys. Lett. 83 (2003) 4981–4983.[13] X.B. Yan, L. Gottardo, S. Bernard, P. Dibandjo, A. Brioude, H. Moutaabbid, P.

Miele, Chem. Mater. 20 (2008) 6325–6334.[14] J. Yan, A.J. Wang, D.P. Kim, Micropor. Mesopor. Mater. 100 (2007) 128–133.[15] Q.D. Nghiem, S.J. Cho, D.P. Kim, J. Mater. Chem. 16 (2006) 558–562.[16] C. Vakifahmetoglu, P. Colombo, Adv. Eng. Mater. 10 (2008) 256–259.[17] J.H. Eom, Y.W. Kim, J. Am. Ceram. Soc. 93 (2010) 2463–2466.[18] L. Ye, Z.H. Ji, W.J. Han, J.D. Hu, T. Zhao, J. Am. Ceram. Soc. 93 (2010) 1156–1163.[19] (a) L. Biasetto, P. Colombo, M.D.M. Innocentini, S. Mullens, Ind. Eng. Chem. Res.

46 (2007) 3366–3372;(b) S. Costacurta, L. Biasetto, E. Pippel, J. Woltersdorf, P. Colombo, J. Am.Ceram. Soc. 90 (2007) 2172–2177.

[20] C. Vakifahmetoglu, P. Colombo, A. Pauletti, C. Fernandez, Martin, F.Babonneau, Int. J. Appl. Ceram. Technol. 7 (2010) 528–535.

[21] P. Krawiec, D. Geiger, S. Kaskel, Chem. Commun. 23 (2006) 2469–2470.[22] J. Yan, A.J. Wang, D.P. Kim, J. Phys. Chem. B 110 (2006) 5429–5433.[23] Y.D. Blum, S.M. Johnson, US Pat. 5 635 (1997) 250.[24] J.D. Torrey, R.K. Bordia, C.H. Henager Jr, Y. Blum, Y. Shin, W.D. Samuels, J.

Mater. Sci. 41 (2006) 4617–4622.[25] D.Y. Zhao, J.L. Feng, Q.S. Huo, N. Melosh, G.H. Fredrickson, B.F. Chmelka, G.D.

Stucky, Science 279 (1998) 548–552.[26] R. Ryoo, S.H. Joo, S. Jun, J. Phys. Chem. 103 (1999) 7743–7746.[27] A.H. Lu, W. Schmidt, A. Taguchi, B. Spliethoff, B. Tesche, F. Schüth, Angew.

Chem. Int. Ed. 41 (2002) 3489–3492.[28] H. Fukui, H. Ohsuka, T. Hino, K. Kanamura, Applied Materials Interfaces 2

(2010) 998–1008.[29] X.D. Zhang, X. X Huang, G.W. Wen, X. Geng, J.D. Zhu, T. Zhang, H.W. Bai,

Nanotechnology 21 (2010) 385601–385608.[30] A. Karakuscu, R. Guider, L. Pavesi, G.D. Soraru, J. Am. Ceram. Soc. 92 (2009)

2969–2974.