An Eco-friendly Soft Template Synthesis of Mesostructured Silica-Carbon Nanocomposites for Acid Catalysis

Article abstract of DOI:10.1002/cctc.201500728

The synthesis of ordered mesoporous silica-carbon composites was explored by employing TEOS and sucrose as the silica and carbon precursor respectively, and the triblock copolymer F127 as a structure-directing agent via an evaporation-induced self-assembly (EISA) process. It is demonstrated that the synthesis procedures allow for control of the textural properties and final composition of these silica-carbon nanocomposites via adjustment of the effective SiO₂/C weight ratio. Characterization by SAXS, N₂ physisorption, HRTEM, TGA, and ¹³C and ²⁹Si solid-state MAS NMR show a 2D hexagonal mesostructure with uniform large pore size ranging from 5.2 to 7.6 nm, comprising of separate carbon phases in a continuous silica phase. Ordered mesoporous silica and non-ordered porous carbon can be obtained by combustion of the pyrolyzed nanocomposites in air or etching with HF solution, respectively. Sulfonic acid groups can be readily introduced to such kind of silica-carbon nanocomposites by a standard sulfonation procedure with concentrated sulfuric acid. Excellent acid-catalytic activities and selectivities for the dimerization of styrene to produce 1,3-diphenyl-1-butene and dimerization of α-methylstyrene to unsaturated dimers were demonstrated with the sulfonated materials. Tintin and the sulfurous nanocatalysts: Sulfonic acid groups can be readily introduced to mesostructured silica-carbon nanocatalysts to reveal excellent acid-catalytic activities and selectivities for the dimerization of styrene to produce 1,3-diphenyl-1-butene and the dimerization of α-methylstyrene.

Full text of DOI:10.1002/cctc.201500728

DOI: 10.1002/cctc.201500728
Full Papers
An Eco-friendly Soft Template Synthesis of Mesostructured
Silica-Carbon Nanocomposites for Acid Catalysis
Ruyi Zhong,[a] Li Peng,*[a] Filip de Clippel,[a] Cedric Gommes,[b] Bart Goderis,[c] Xiaoxing Ke,[d]
Gustaaf Van Tendeloo,[d] Pierre A. Jacobs,[a] and Bert F. Sels*[a]
The synthesis of ordered mesoporous silica-carbon composites
was explored by employing TEOS and sucrose as the silica and
carbon precursor respectively, and the triblock copolymer F127
as a structure-directing agent via an evaporation-induced self-
assembly (EISA) process. It is demonstrated that the synthesis
procedures allow for control of the textural properties and
final composition of these silica-carbon nanocomposites via
adjustment of the effective SiO2/C weight ratio. Characteriza-
tion by SAXS, N2 physisorption, HRTEM, TGA, and 13C and 29Si
solid-state MAS NMR show a 2D hexagonal mesostructure with
uniform large pore size ranging from 5.2 to 7.6 nm, comprising
of separate carbon phases in a continuous silica phase. Or-
dered mesoporous silica and non-ordered porous carbon can
be obtained by combustion of the pyrolyzed nanocomposites
in air or etching with HF solution, respectively. Sulfonic acid
groups can be readily introduced to such kind of silica-carbon
nanocomposites by a standard sulfonation procedure with
concentrated sulfuric acid. Excellent acid-catalytic activities and
selectivities for the dimerization of styrene to produce 1,3-di-
phenyl-1-butene and dimerization of a-methylstyrene to unsa-
turated dimers were demonstrated with the sulfonated
materials.
Introduction
Since the discovery of ordered mesoporous silica materials in
1990s, synthesis and applications of mesoporous solids have
received intensive attention, owing to their wide applications
in catalysis, adsorption, separation, devices, and drug delivery,
etc.[1–7] With well-developed soft-template and hard-template
(nanocasting) methods, mesoporous materials based on vari-
ous compositions have been prepared, such as silica, heteroa-
tom doped silica, organosilica, metal oxide, metal, polymer,
carbon as well as composite.[8–12] Among these materials, mes-
oporous silica-carbon composites are an attractive family of
novel mesoporous materials. Besides their well-controlled pore
structures, high surface areas, and large and tunable pore sizes
which facilitate the diffusion of reactants and products inside
the pores, they combine the properties of inorganic compo-
nents, such as good mechanical and thermal stability, as well
as organic components, such as hydrophobicity and versatility
for further functionalization.[1] These materials have, conse-
quently, received significant attention recently for their poten-
tial applications in catalysis, adsorption, separation, optics devi-
ces, and electrodes.[13–23] Ordered mesoporous silica-carbon
composites prepared by infiltration of carbon precursors into
preformed mesoporous silica followed by carbonization.[14–20]
However, this procedure is laborious and time consuming,
owing to the involvement of the pre-formation of the ordered
mesoporous silica template. Mesoporous silica-carbon compo-
sites can also be prepared by carbonization of mesoporous or-
ganosilica/surfactant mesophases, or mesoporous silica/surfac-
tant mesophases.[24–28] One drawback is that organosilanes are
expensive and difficult to obtain and another drawback is that
organic functional groups or surfactant may block the pores.
Recently, a multi-component assembly approach has been
used to prepare ordered mesoporous silica-carbon nanocom-
posites by using phenolic resins (resols) as carbon precursor,
TEOS as silica precursor, and triblock copolymer F127 as a tem-
plate.[29,30] The successful synthesis is mainly attributed to the
unique characteristics of the thermosetting phenolic resins for
the formation of three-dimensional (3D) cross-linked networks,
such as abundant benzyl and phenolic hydroxyl groups, and
controllable polymerization rate.[31] However, resols are hazard-
ous as they consist of derivatives of phenols and formaldehyde
which are all carcinogenic. Therefore, it is interesting to devel-
[a] R. Zhong, L. Peng, F. de Clippel, P. A. Jacobs, B. F. Sels
Centre for Surface Science and Catalysis
KU Leuven
Kasteelpark Arenberg 23 - bus 2461, 3001 Heverlee (Belgium)
Fax: (+32)016-321998
E-mail: lisa.lipeng@gmail.com
bert.sels@biw.kuleuven.be
[b] C. Gommes
Department of Chemical Engineering
University of Liege
Building B6A, 4000 Liege (Belgium)
[c] B. Goderis
Polymer Chemistry and Materials
KU Leuven
Celestijnenlaan 200f - bus 2404, 3001 Heverlee (Belgium)
[d] X. Ke, G. V. Tendeloo
Electron Microscopy for Materials Science
Universiteit Antwerpen
Groenenborgerlaan 171, 2020 Antwerpen (Belgium)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cctc.201500728.
This publication is part of a Special Issue on “Carbon in the Catalysis
Community”. Once the full issue has been assembled, a link to its Table
of Contents will appear here.
ChemCatChem 2015, 7, 3047 – 3058
3047
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
op a novel and eco-friendly carbon precursor which
can replace resols in the synthesis.
Here we explore the synthesis of novel ordered
mesoporous silica-carbon nanocomposite using
a simple and efficient synthesis through multi-com-
ponent assembly by using TEOS, sucrose as precur-
sors and F127 as a template. Sucrose was chosen as
the carbon precursor given following considerations.
Firstly, sucrose is a readily-available natural product,
which will make this process easy and green. Second-
ly, the multiple adjacent hydroxyl groups of sucrose
probably promote the formation of hydrogen bond-
ing with the polar PEO blocks of F127 and silica olig-
omers, which makes the tri-constituent co-assembly
possible. Thirdly, sulfonic acid groups (SO3H) can be
readily introduced to incompletely carbonized su-
crose32 which gives rise to potential application of
such silica-carbon nanocomposites as solid-acid
catalysts.
Figure 1. SAXS patterns of: A) silica-carbon nanocomposites samples before a) Si33C66,
b) Si50C50, c) Si66C33, and after pyrolysis d) Si33C66-1073, e) Si50C50-1073, f) Si66C33-1073; and
after removal of carbon or silica from the pyrolyzed composites to obtain pure B) silica
or C) carbon samples, i), ii), and iii) corresponding to their mother composites of Si33C66,
Si50C50, and Si66C33, respectively.
Sucrose has been widely used as carbon precursor for the
synthesis of mesoporous carbon through a hard-template
method.[33,34] However, to the authors’ knowledge, sucrose has
seldom been used as carbon precursor for the direct soft-tem-
plate synthesis of ordered mesoporous structure. Herein, evap-
oration-induced co-assembly of TEOS, sucrose and F127 for
the synthesis of mesoporous silica-carbon nanocomposites
was extensively investigated. These nanocomposites possess
an ordered 2D hexagonal mesostructure, high surface areas
(120–500 m2 gÀ1), and uniform large pore sizes (5.2–7.6 nm).
The textural properties and compositions can be easily tuned
by adjusting the initial mass ratios of TEOS to sucrose and the
pyrolysis temperature, which were further demonstrated to
greatly influence the catalytic performance. The sulfonated
silica-carbon nanocomposites were applied to the dimerization
of styrene to produce linear unsaturated dimer, 1,3-diphenyl-1-
butene (DPB) and the dimerization of a-methylstyrene (AMS)
to produce acyclic dimers. These unsaturated dimers are indus-
trially important as chain transfer agents. Moreover, the
carbon-bound sulfonic acid groups will show high accessibility
to such hydrophobic substrate molecules due to the apolar
nature of the carbon component and the mesoporosity of the
nanocomposite.
(Figure 1A, a–c) exhibit a main scattering peak at a vector of
q* and other dim peaks for some samples at larger vectors.
p
These dim peaks are located at vectors of 3/2 q* and 2q*,
and characteristic of an ordered 2D hexagonal mesostructure
with the p6mm symmetry. The ordered domains are of a rela-
tively limited size, which in turn causes the blurriness of the
secondary scattering peaks due to the Scherrer line broaden-
ing effect. The average distance between neighboring pores
p
was calculated by using the formula a=4 p/( 3 q*), and listed
in Table 1.
After pyrolysis under N2 at 1073 K, the silica-carbon nano-
composites show SAXS scattering peaks (Figure 1A, d–f), as-
signed to the (100) reflection of the hexagonal p6mm meso-
structure. The pore distance was shortened compared to the
parent silica-carbon nanocomposite, owing to the transforma-
tion of sucrose into carbonization products and the decompo-
sition of the template F127. The shrinkage degree decreased
with the increasing silica/carbon ratio, which demonstrates the
stabilization effect of the silica component for the whole meso-
structure during thermal treatment. The final pore distance of
Si33C66-1073 was 12.4 nm, slightly larger than Si50C50-1073
(11.3 nm) and Si66C33-1073 (11.5 nm). Furthermore, the scatter-
ing peak of Si33C66-1073 was much weaker than that of Si50C50-
1073 and Si66C33-1073, and the reduced contrast may be attrib-
utable to the filling of mesopores by carbon. Typical TEM
images of silica-carbon nanocomposite and its derived pure
silica and carbon materials, Si50C50-1073, Si50C50-Si and Si50C50-C
are shown in Figure 2.
Results and Discussion
Synthesis of ordered mesoporous silica-carbon composites
Mesoporous silica-carbon nanocomposites are prepared from
an EISA method by using triblock copolymer F127
(EO106PO70EO106) as a template, ethanol and H2O as solvent,
TEOS and sucrose as silica and carbon precursors, respective-
ly.[35] By adjusting the initial mass ratios of TEOS and sucrose,
Si33C66, Si50C50, and Si66C33 nanocomposites with different silica
and carbon content can be obtained. All these as-synthesized
samples from EISA processes are light yellow to deep yellow
membranes without obvious macrophase separation. SAXS
patterns of the as-synthesized silica-carbon nanocomposites
Uniform hexagonally arranged channel arrays are observed
in large domains in Si50C50-1073 (Figure 2A), clearly showing
a highly ordered hexagonal mesostructure. The average dis-
tance between neighboring pores determined from the TEM
image was 10.9 nm, consistent with the value of 11.3 nm from
its SAXS pattern. The average pore size was 6.0 nm and the
pore-wall thickness was 5.3 nm for Si50C50-1073 according to its
TEM image.
N2 sorption isotherms and pore size distributions of SimCn-
1073 nanocomposites are shown in Figure 3A. The corre-
ChemCatChem 2015, 7, 3047 – 3058
www.chemcatchem.org
3048
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Pure silica and carbon materials from silica-carbon
composites
Table 1. Composite and physicochemical properties of carbon/silica nanocomposites
and of the corresponding silica and carbon frameworks obtained after the removal of
carbon and silica, respectively.[a]
Mesoporous silica materials were obtained from the
silica-carbon nanocomposites after calcination in
air to remove the carbon component. The SAXS pat-
terns of pure silica materials (Figure 1B) show one
scattering peak which is assigned to a hexagonally
ordered mesostructure. The calculated average dis-
tances between neighboring pores, listed in Table 1,
are close to the values of their parent nanocompo-
sites. Therefore, the framework of silica was almost
kept intact during the calcination process and was
always of p6mm symmetry, despite the difference in
carbon content between the nanocomposites. The
TEM image of Si50C50-Si (Figure 2B) resembles well
that of Si50C50-1073, which implies that the carbon
phase is dispersed homogeneously in the silica
framework.
Sample
C
a
SBET
Vtot
Vmeso
Vmicro
D
[nm]
[%] [nm] [m2 gÀ1
]
[cm3 gÀ1
]
[cm3 gÀ1
]
[cm3 gÀ1
]
Si33C66
Si50C50
Si66C33
Si33C66-1073
Si50C50-1073
Si66C33-1073
Si33C66-Si
Si50C50-Si
Si66C33-Si
Si33C66-C
Si50C50-C
Si66C33-C
Si33C66-673-SO3H
Si50C50-673-SO3H
Si66C33-673-SO3H
Si33C66-823-SO3H
Si50C50-823-SO3H
Si66C33-823-SO3H
–
–
–
16.8
15.3
13.9
–
–
–
–
–
–
–
–
–
–
–
–
49 12.4
32 11.3
19 11.5
155
177
156
188
410
381
794
1118
1086
471
424
493
–
0.06
0.20
0.22
0.12
0.36
0.41
0.39
0.82
0.69
0.20
0.25
0.44
–
0.01
0.16
0.20
0.12
0.33
0.38
0.15
0.75
0.59
0.02
0.13
0.33
–
0.05
0.04
0.02
0.0
0.03
0.03
0.04
0.07
0.10
0.18
0.12
0.11
–
5.2
6.3
6.4
2.6, 5.0
2.6, 6.2
2.6, 6.2
–
3.5
2.6
–
7.6
6.9
–
6.9
6.7
0
0
0
11.7
10.6
11.4
–
100
100 10.5
100
62 13.6
49 12.7
30 12.7
54 12.8
38 11.9
23 11.9
–
124
332
0.10
0.39
0.07
0.35
0.03
0.04
N2 sorption isotherms of Si50C50-Si and Si66C33-Si
(Figure 3B, up) are type IV curves with a H2-type hys-
teresis loop. The steep capillary condensation also
mainly occurred at a relative pressure range of 0.60–
0.75, corresponding to uniform mesopores of 5.5–
7.5 nm similar to those in their parent silica-carbon
nanocomposites. Combined with the SAXS results,
these mesopores originate from the F127 micell or-
ganization in the original structure and are opened
by thermally decomposing the F127 molecules. How-
[a] C% were the mass percentage of carbon in the dry samples, determined from
TGA; Surface area (SBET) is calculated by the BET method; Total pore volume (Vtot) was
calculated from the saturation plateau at high relative pressures; Mesopore volume
(Vmeso) was calculated after subtraction of the micropore volume (Vmicro) as obtained
by the t-plot method; Pore size (D) at the maxima of pore size distribution was calcu-
lated from adsorption branches of the isotherms, based on the BJH model.
sponding values of BET surface area, total pore volume, meso-
pore volume and average mesopore size are presented in
Table 1. Si50C50-1073 and Si66C33-1073 show a type IV curve with
a H1-type hysteresis loop, typical of a well-ordered cylindrical
mesoporous structure. Their isotherms show distinct steep ca-
pillary condensation steps at P/P0 =0.60–0.75, corresponding
to uniform mesopores. The pore size of Si50C50-1073 was
6.3 nm with a narrow distribution and the pore-wall thickness
was 5.0 nm, which is coincident with the TEM measurement.
The mesopore volumes of Si50C50-1073 and Si66C33-1073 were
0.20 and 0.22 cm3 gÀ1 respectively, predominant in the total po-
rosity. These silica-carbon nanocomposites mainly contain 5.5–
7.5 nm mesopores with silica and carbon components coexist-
ing in the pore-walls, whereas the carbon-rich Si33C66-1073
shows a type IV isotherms with a H1 +H3 mixed hysteresis
loop. The total pore volume was only 0.06 cm3 gÀ1 and the
mesoporosity was hardly present (Vmeso =0.01 cm3 gÀ1). Howev-
er, the SAXS pattern of Si33C66-1073 (Figure 1A, a) reveals its
hexagonal mesostructure. As determined from the TGA analy-
sis (Figure S1), carbon content in Si33C66-1073 was 49%, a lot
higher than in Si50C50-1073 (32%) and Si66C33-1073 (19%). It is,
therefore, very likely that excessive part of the carbon phase in
Si33C66-1073 located inside the silica-carbon mesopores and
caused the closure of most mesopores. The carbon in the mes-
opores was probably in the form of aggregates of small platy
carbon sheets, forming slit-shaped pores, causing the H3-type
hysteresis loop.
ever, the hysteresis loops of pure silica materials were partially
deformed, with regard to the ideal parallel H1-type loops of
their parent silica-carbon nanocomposites, probably due to the
presence of secondary mesopores, which was generated
during the burn-off of carbon in the pore-walls. The pore size
distributions of Si50C50-Si and Si66C33-Si (Figure 3B, bottom)
were indeed bimodal, showing pore sizes centered at 2.6 nm
in addition to the primary pore size of 6.2 nm. These results in-
dicate that the carbon phase in Si50C50-1073 and Si66C33-1073 is
located in small domains of mean size of 2.6 nm in the pore-
walls.
N2 sorption isotherm of Si33C66-Si (Figure 3B, up) is type IV
curve with a H2-type hysteresis loop, significantly different
from its parent Si33C66-1073. The opening of mesopores which
were previously filled by amorphous carbon, accounts for the
large H2-type hysteresis loop observed on the pure silica mate-
rial and the simultaneously elimination of the H3-type loop of
the intrapore carbon sheets. Bimodal pore sizes are also ob-
served for the Si33C66-Si (Figure 3B, bottom). The smaller meso-
pores were ꢀ2.6 nm, the same as Si50C50-Si and Si66C33-Si,
which indicates similar nanosized carbon domains located
inside the composite pore-walls of Si33C66-1073. The BET surface
area, the total porosity and the mesoporosity of these three
pure silica materials all increased compared to their parent
silica-carbon nanocomposites (Table 1, entries 4–9), also evi-
dencing that the carbon component forms small carbon
sheets and is dispersed within the pore-walls in the nanocom-
posites. After removal of carbon component, mesoporous silica
ChemCatChem 2015, 7, 3047 – 3058
www.chemcatchem.org
3049
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
amorphous carbon inside the F127-derived big mesopores.
These results show that carbon is not well organized in a con-
tinuous phase throughout the materials, when compared to
silica, leading to a loss of pore structure upon removal of silica.
N2 sorption isotherms of all these three derivative carbon
materials (Figure 3C, top) are of type-IV with a H2 +H3-type
hysteresis loop, consistent with a disordered pore structure.
Etching of silica component would result in lots of voids in the
pore-walls and largely reduce the mesoscopic regularity. These
small mesopores resulted from silica removal coincided well
with the left-shifted capillary condensation steps in a relative
pressure range of 0.45-0.65, in relative to the parent silica-
carbon nanocomposites. Meanwhile, the BET surface area
(794–1118 m2 gÀ1) and mesopore volume (0.15–0.75 cm3 gÀ1)
(Table 1) were increased after removal of silica component,
also due to the newly generated small mesopores (Figure 3C,
bottom) in the pore-walls. The mesoporosity of Si33C66-C was
still much smaller than Si50C50-C and Si66C33-C as a result of the
microporous carbon aggregates in the F127-derived large
mesopores in Si33C66-1073.
Formation of mesostructured silica-carbon framework
We propose that the formation of composites undergoes a tri-
constituent assembly process of organic sucrose precursor, sili-
cate oligomer generated through the hydrolysis of TEOS, and
amphiphilic triblock copolymer F127 (Scheme 1). F127, sucrose
and TEOS are well dissolved in acidic EtOH/H2O solution. TEOS
is hydrolyzed rapidly in presence of HCl to form silicate oligo-
mers. With the evaporation of EtOH solvent at 313 K, the tri-
block copolymer F127 tends to aggregate into columnar mi-
celles with hydrophobic PPO block as a core surrounded by
a polar PEO block shell. Simultaneously, sucrose and the silicate
oligomers, both possessing plenty of hydroxyl groups, interact
strongly with the long EO chains of the copolymer via hydro-
gen-bonding. Sucrose and the silicate oligomers are thus ag-
gregated around the EO domains of the F127 micelles. Mean-
while, under acidic conditions, PEO blocks, as well as OH
groups of silica oligomers and sucrose, can be partially proton-
ated and can self-assemble together through the electrostatic
interaction with ClÀ anions.[36] By the two assembly mecha-
nisms, continuous evaporation of ethanol promotes the con-
struction of hexagonally mesostructured composites
(Scheme 1A) of F127 micelles with sucrose and silicate oligo-
mers. The condensation and cross-linking of the silicate oligo-
mers to form SiO2 framework proceeds simultaneously, where-
as sucrose does not polymerize at 313 K. Owing to the tetrahe-
dral structure of SiO2, cross-linking of silicate oligomers can
occur around sucrose molecules and incorporate these
“carbon domains” into the SiO2 framework.[29] This can also be
understood from the point that silica oligomers self-assemble
around the secondary non-surfactant template, sucrose,[37] in
the hydrophilic polyethylene oxide (PEO) region of the F127
micelles.
Figure 2. TEM images of the A) silica-carbon nanocomposite Si50C50-1073
and the pure B) silica (Si50C50-Si), and C) carbon (Si50C50-C) samples derived
therefrom after removal of carbon and silica, respectively.
materials can be obtained with a hierarchical organization of
the pore architecture.
Pure carbon materials were obtained from the silica-carbon
nanocomposites by etching of silica component with HF solu-
tion. SAXS patterns for these SimCn-C materials are shown in
Figure 1C. Without the presence of silica component, the or-
dering of the mesostructure was largely impaired. Si50C50-C dis-
played a weak and broad scattering peak, and the scattering
peak of Si66C33-C was even more obscure. However, the calcu-
lated average pore distances for both Si50C50-C and Si66C33-C
was not far from the values for their parent silica-carbon nano-
composites, showing the traces of the original hexagonal mes-
ostructure. The severe degradation of the mesostructure after
silica removal in the composite pore-walls is also revealed by
the TEM image of Si50C50-C (Figure 2C). As for Si33C66-C, no scat-
tering peak at all was observed from its SAXS pattern. This ir-
regular porous Si33C66-C material could be constructed by the
small carbon sheets from the composite pore-walls and those
When sucrose is added to a high amount, like in the case of
Si33C66 (F127/TEOS/sucrose=1:1.3:1.74 in mass ratio), in con-
trast with Si50C50 (F127/TEOS/sucrose=1:1.3:0.87 in mass ratio)
ChemCatChem 2015, 7, 3047 – 3058
www.chemcatchem.org
3050
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
the severe penetrating of su-
crose into the PPO cores of the
copolymer F127 micelles, which
destroys the ordered distribution
of carbon species in the final
product. However, pure silica
mesostructure with 2D hexago-
nal p6mm symmetry can be ob-
tained at the TEOS/F127 mass
ratio ranging from 1 to 2 via
similar EISA process.[39] Therefore,
in our synthesis, the mesostruc-
tural ordering increases with the
increase of silica/carbon ratio
and a too-high amount of su-
crose causes a structural disor-
dering, although the basic
p6mm symmetry of the packing
of the micelles is unchanged.
This is quite different from the
Figure 3. [Top] N2 adsorption (closed symbols)/desorption (open symbols) isotherms and [Bottom] pore size distri-
bution of A) composites, B) pure silica, and C) pure carbon materials derived from: i) Si33C66, ii) Si50C50, and
iii) Si66C33. For clarity, the offset between the isotherms along the y-axis is 100 cm3 gÀ1 in A/B and 200 cm3 gÀ1 in C.
silica-carbon
nanocomposites
Scheme 1. Evaporation-induced triconstituent assembly (EISA) to ordered
Scheme 2. EISA to silica-carbon nanocomposite of high C content (Si33C66),
the corresponding ordered mesoporous silica and porous carbon materials,
and its derived sulfonated silica-carbon composites. Red represents silica,
while green to black represents carbonaceous component.
mesoporous silica-carbon composites, sulfonated mesoporous silica-carbon
composites and the corresponding ordered mesoporous silica and porous
carbon materials in the case of moderate to low C content (Si50C50 and
Si66C33). Red represents silica, while green to black represents carbonaceous
component.
prepared from the phenolic resin-TEOS-F127 system. A hexago-
nal mesostructure without silica component can be obtained
directly from the assembly of phenolic resin around F127 mi-
celles.[29,31] In that case, the mass ratio of silica/phenolic resin
can be tuned arbitrarily without collapsing the mesostructure.
The SiO2-Sucrose-F127 composite was thermopolymerized
at 433 K to produce SiO2-carbonaceous polymer-F127
(Scheme 1B) and further pyrolyzed to produce SimCn-T
(Scheme 1C1 and 1C2). The transformation of the carbon and
silica species during these processes was monitored by 13C CP-
MAS and 29Si solid-state MAS NMR respectively and the results
and Si66C33 (F127/TEOS/sucrose=1:1.3:0.44 in mass ratio), the
excessive part of sucrose molecules can blend with the entire
F127 copolymer and enter the hydrophobic polypropylene
oxide (PPO) cores of the micelles (Scheme 2A). Sucrose acts to-
gether with F127 as structure-directing agent besides as the
carbon precursor when its concentration is higher than a cer-
tain “saturation” value of embedding in the silica network in
the pore-walls. Notably, pure mesoporous carbon cannot be
obtained directly from the co-assembly of sucrose and F127
without the addition of silica source,[38] probably resulting from
ChemCatChem 2015, 7, 3047 – 3058
www.chemcatchem.org
3051
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
chemical shifts, 170 and 152 ppm, besides the aromatic carbon
peak of Si50C50-673-SO3H would result from the relatively low
pyrolysis temperature applied. The peaks at 170 and 152 ppm
are assigned to COOH groups and OH-substituted carbons of
phenols respectively.[20,40] The presence of oxygen-containing
species in the carbon phase in the silica-carbon nanocompo-
sites may impact the surface properties and in turn the catalyt-
ic performance. When the pyrolysis temperature was elevated
to 823 K, the peak for polycyclic aromatic carbon grew stron-
ger but the peaks for carbons binding with oxygen atoms
became weaker, indicating the growth and aggregation of
carbon sheets in the silica-carbon nanocomposites during py-
rolysis in N2. By further raising the pyrolysis temperature to
1073 K, the oxygen-containing carbon species were eliminated
with the exclusively one peak at 128 ppm left. Moreover, this
peak decreased largely in intensity which probably results
from a high graphitization degree in Si50C50-1073.
Figure 4. A) 13C and B) 29Si CP-MAS NMR spectra of (blue) Si66C33, (light red)
Si50C50-673-SO3H, (dark red) Si50C50-823-SO3H, (grey) Si50C50-823, and (black)
Si50C50-1073.
The 29Si solid-state MAS NMR spectra of silica-carbon nano-
composites before and after pyrolysis are shown in Figure 4B.
Three peaks at À91, À101 and À110 ppm, which correspond
to Q2 (Si(OSi)2(OH)2), Q3 (Si(OSi)3(OH)) and Q4 (Si(OSi)4) type of
silica respectively, are observed in the 29Si NMR spectrum of
are shown in Figure 4. In the 13C NMR spectrum of the as-syn-
thesized Si66C33 (Figure 4A), peaks at ꢀ170 ppm and in the
range of 110–90 ppm, which represents COOH groups and C= the as-synthesized Si66C33. After pyrolysis, Q2 and Q3 contents
C of aromatics respectively, indicate that sucrose is primarily
polymerized at 433 K. Carbonaceous polymers may be formed
from sucrose by its dehydration into furan-like derivatives and
further condensation with the hydroxylmethylene CH2OH
groups. The CÀO groups in such humin-like chemical struc-
tures should correspond to the broad 13C NMR peak in the
range of 90–55 ppm. Notably, there are two distinctively sharp
peaks, at 70 and 60 ppm respectively, overlapping with the
broad peak. They are ascribed to CHÀO and CH2ÀO groups of
F127 respectively, together with the peak at ꢀ20 ppm, which
represents aliphatic carbon, confirming the presence of F127
in the as-synthesized nanocomposites. The sharp signal from
ether structures of F127 in contrast with the broad signal from
those of sucrose-derived saccharoidal-like polymers, implies
a higher mobility of the carbon of F127, which strongly sup-
ports the proposed image that F127 forms relatively independ-
ent hexagonally-arranged arrays but sucrose is closely entan-
gled in the silica framework in the stage of EISA.
In the 13C NMR spectra of Si50C50 after pyrolysis at 673, 823,
or 1073 K (Figure 4A), the peak at ꢀ20 ppm for aliphatic
carbon, as well as the sharp peaks in the range of 90–55 ppm
disappeared, which indicates the removal of F127 template
during pyrolysis at temperatures higher than or equal to 673 K.
The desorption and decomposition of F127 would create po-
rosity in the nanocomposites (Scheme 1C1 and 1C2). Simulta-
neously, the broad peak for CÀO in sucrose-derived polymers
disappeared whereas the peak at 128 ppm became dominant.
This peak is assigned to aromatic carbon, characteristic of
carbon materials from pyrolysis.[20,40] Indeed, sucrose molecules
start to form polycyclic aromatic carbon sheets for pyrolysis
temperatures above 573 K by further converting furan rings to
benzene rings.[32,41] Since sulfonation would not affect the
13C NMR spectra as indicated by the very similar patterns of
Si50C50-823 and Si50C50-823-SO3H, the minor peaks at higher
decreased significantly while Q4 content increased. With the
increase of pyrolysis temperature, Q3/O4 decreased to a larger
extent. In the Si50C50-1073, the Q2 and Q3 types of silica almost
disappeared with only Q4 silica species left. These results indi-
cate that during the pyrolysis, silica were continuously solidi-
fied by condensation and cross-linking. A rigid silica framework
can be obtained by pyrolysis at a high temperature.
Furthermore, the 13C CP-MAS and 29Si solid-state MAS NMR
spectra of these pyrolyzed silica-carbon nanocomposites show
no indication of the formation of Si-C species (Figure 4).
Hence, it is plausible that the structure of silica-carbon nano-
composite would be superposition of the structures of its de-
rived pure silica and carbon materials by removal of carbon
and silica component respectively. On account of the rigidity
of the silica framework, pure silica materials SimCn-Si
(Scheme 1D and 2D) after the burn-off of carbon component
show an unchanged mesostructure with newly generated
ꢀ2.6 nm mesopores in addition to the primary ꢀ6.2 nm meso-
pores, compared to their parent silica-carbon nanocomposite
SimCn-1073. Nevertheless, pure carbon materials SimCn-C
(Scheme 1E) after etching of silica component show a deterio-
rated ordering of mesostructure with a large amount of small
mesopores (<4 nm). Hence, the parent silica-carbon nanocom-
posites possess a “homogeneous” framework constituted by
nanosized silica and carbon components interpenetrated with
each other. For Si33C66 where sucrose is in a too-high amount,
the pyrolyzed material is microporous as most of the large
mesopores are filled with amorphous carbon sheets from ex-
cessive sucrose penetrated into the PPO micell cores
(Scheme 2C1 and 2C2).
ChemCatChem 2015, 7, 3047 – 3058
www.chemcatchem.org
3052
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Sulfonation of the mesostructured silica-carbon nanocompo-
sites and their acid properties
ture, the intensity of the stretching vibration band of C=O de-
creases, but the intensity of the stretching vibration band of
C=C increases, as a result of the increased carbonization
degree. However, the stretching bands of SO3H cannot be dif-
ferentiated for the overlapping with the strong bands of SiÀO
stretching. In the FTIR spectra of HF solution-treated SimCn-823-
SO3H samples (Figure S6)†, the bands at 1032 and 1007 cmÀ1
for the stretching of SO3H groups are visible, showing the suc-
cessful functionalization of these silica-carbon nanocomposites
with SO3H groups. The density of SO3H groups was determined
by titration and correlated with the carbon content of the
sample as show in Figure S7. SO3H site density increases linear-
ly with the carbon content at the respective pyrolysis tempera-
ture, which indicates that similar carbonization degree of su-
crose is obtained, thus showing similar reactivity towards
H2SO4. However, the slope of the fitting line for the pyrolysis
temperature of 823 K is smaller than the one for 673 K, indicat-
ing that larger carbon sheets formed at higher pyrolysis tem-
perature can be closely packed and with less oxygen-contain-
ing groups,[40] and therefore more difficult to be sulfonated.
The 31P NMR spectra obtained after an adequate amount of
TMPO adsorption and evacuation on these sulfonated silica-
carbon nanocomposites are shown in Figure 5. The dashed
curves indicate Gaussian deconvolution of the spectra. The ab-
sence of resonance peaks at 41 ppm for all the 31P NMR spec-
tra indicates that no excess of TMPO aggregates on the surface
and is a proof of sufficient evacuation.[40,43] The resonance
peaks can be divided into three categories. Peaks at 78–
To evaluate the catalytic potential of these novel mesoporous
silica-carbon nanocomposites, the as-synthesized SimCn nano-
composites were pyrolyzed at a moderate temperature (673 or
823 K) and functionalized with sulfonic acid groups. The
unique characteristics of silica-carbon nanocomposites, due to
the combination of well-ordered mesostructure of silica and
easy functionalization of carbon component, are promising to
fabricate such solid-acid catalysts with a high amount of easily
accessible SO3H groups. It is known that SO3H can be intro-
duced into the incompletely carbonized sugar by sulfonation
with concentrated H2SO4.[32] Meanwhile, silica component serve
as a rigid support to maintain the ordered mesostructure
during the harsh sulfonation treatment.[33,42]
The textural properties of these sulfonated silica-carbon
nanocomposites were studied with SAXS and N2 sorption
measurements (Figure S3 and S4). The results are summarized
in Table 1, entries 13–18. Sulfonation would hardly influence
the pore structure, except for slight loss of carbonaceous spe-
cies, owing to side-reactions such as oxidation and dehydra-
tion during sulfonation.[42] Therefore, the textural parameters
of SimCn-673/823-SO3H are similar to those of SimCn-673/823,
which differ from SimCn-1073 merely by the lowering of the py-
rolysis temperature. Since there is no severe degradation of
mesostructure ordering even when pyrolysis was performed at
1073 K, the mesostructure of SimCn-673/823-SO3H sample
would be more or less the same with the above-analyzed
structure of SimCn-1073. The SAXS patterns of SimCn-673/823-
SO3H samples (Figure S3) all show a scattering peak, assigned
to the (100) reflection of the hexagonal mesostructure. The
calculated pore distance is consistent with the rule that slight
shrinkage of the framework occurs during pyrolysis. Among
samples pyrolyzed at the same temperature, the scattering
peak of Si33C66-673/823-SO3H is the weakest and corresponding
to a slightly larger pore distance than Si50C50-673/823-SO3H and
Si66C33-673/823-SO3H. These results also imply that excessive
sucrose in the synthesis of Si33C66 samples can penetrate into
the hydrophobic domain of the F127 micelles and expand the
pore structure, but impair the ordering of the mesostructure at
the same time. The N2 sorption isotherms of Si50C50-673/823-
SO3H and Si66C33-673/823-SO3H samples (Figure S4A) show
a typical type IV curve with a H1-type hysteresis loop. They are
well-ordered cylindrical mesoporous materials, with a narrow
pore size distribution centered at ꢀ7 nm (Figure S4B). On the
contrary, Si33C66-673-SO3H is microporous and Si33C66-823-SO3H
has no porous structure at all (Figure S4), as a result from the
excess of carbon in the F127-derived mesopores. Notably, the
samples pyrolyzed at 673 K all possess high micropore volume
with respect to those pyrolyzed at higher temperatures, per-
haps due to the loose packing and incomplete formation of
carbon sheets at 673 K.
The acid properties of these sulfonated silica-carbon nano-
composites were investigated by FTIR and TMPO probed
31P NMR spectroscopy. The FTIR spectra of SimCn-673/823-SO3H
(Figure S5) show that with the increase of pyrolysis tempera-
Figure 5. 31P NMR spectra of catalysts after adsorption of trimethylphosphine
oxide (TMPO): A1) Si33C66-673-SO3H, A2) Si33C66-823-SO3H, B1) Si50C50-673-
SO3H, B2) Si50C50-823-SO3H, C1) Si66C33-673-SO3H, and C2) Si66C33-823-SO3H.
ChemCatChem 2015, 7, 3047 – 3058
www.chemcatchem.org
3053
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
86 ppm are ascribed to TMPO adsorbed on strong acid sites,
i.e., SO3H groups in these sulfonated materials, in consistency
with the previous assignments of TMPO-probed 31P MAS NMR
spectra of sulfonated carbon.[20,40] Peaks at lower chemical
shifts, ꢀ75, 65, 54 ppm can be assigned to TMPO adsorbed on
weak acid sites like COOH, phenolic OH and SiÀOH. Peaks at
48 ppm are ascribed to physisorbed TMPO. For samples pyro-
lyzed at 673 K (Figure 5A1, 5B1, and 5C1), similar high chemi-
cal shifts for TMPO on SO3H groups are observed, which indi-
cates that similar acid strength is obtained after sulfonation of
the carbonaceous phase. The areal percentage of this peak in
the total 31P resonance peaks decreases from 43 to 32 to 25%
for Si33C66-673-SO3H, Si50C50-673-SO3H, and Si66C33-673-SO3H.
These results are in line with the SO3H group densities deter-
mined by titration. However, an opposite trend to the results
of titration is found for the 31P resonance peak areas, corre-
sponding to SO3H groups on the 823 K-pyrolyzed samples (Fig-
ure 5A2, 5B2, and 5C2). The area percentage of this peak in-
creases from 0 to 17 to 32% with the decrease of carbon con-
tent, whereas the titration data showed a decrease of acidity
with decreasing carbon content. This phenomenon is likely
due to significant structural restriction around the strong acid
sites on these 823 K-pyrolyzed samples. By referring to the tex-
tural properties of these sulfonated silica-carbon composites
(Table 1, entries 13–18), the decreased mesoporous ordering
with the increased carbon content in the pore structure is re-
garded as the major cause of a reduced accessibility of TMPO
molecules to the SO3H groups.
Performance in acid catalysis
The catalytic performance of sulfonated silica-carbon nano-
composites was demonstrated through dimerization of styrene
and dimerization of a-methylstyrene. The reaction Scheme of
dimerization of styrene is shown in Scheme S1. Protonated sty-
rene reacts with another styrene monomer to form a dimeric
carbocation intermediate which either deprotonates into
a head-to-tail dimer 1,3-diphenyl-1-butene (DPB) or into 1-
methyl-3-phenylindan (MPI) by intramolecular Friedel–Crafts al-
kylation which requires relatively strong acid sites.[44] Small
amounts of trimers and adducts with toluene (the solvent)
were also detected as side products.
Figure 6. Dimerization of styrene over sulfonated silica-carbon nanocompo-
~
&
*
*
)
sites. ( ) Si33C66-673-SO3H, ( ) Si50C50-673-SO3H, ( ) Si66C33-673-SO3H, (
Si50C50-823-SO3H, (~) Si66C33-823-SO3H, (”) CBV720, and ( ) Nafion NR50. Re-
action conditions: styrene (0.50 g), catalyst (50 mg), toluene (5.00 g), reaction
temperature 373 K.
^
tivity for styrene, despite it bears a high amount of SO3H sites
in micropores which are accessible to TMPO molecules (Fig-
ure 5A1). Hence, accessible (effective) sulfonic acid groups to
styrene molecules are entailed for the dimerization of styrene.
Nafion NR50 shows the highest activity of styrene conver-
sion. However, the selectivity towards the linear unsaturated
dimer, DPB, is much lower than all the sulfonated silica-carbon
nanocomposites and higher amounts of cyclic indanic dimer,
MPI and adducts of styrene with toluene were produced. Such
relatively low selectivity of DPB is also observed with CBV720.
It is very possible that the Brønsted acidities of Nafion NR50
and CBV720 are so strong that not only catalyze the equilibri-
um reaction of the protonated dimer intermediate to DPB, but
also catalyze the intramolecular and intermolecular electrophil-
ic substitution reactions. In comparison, the selectivities of DPB
over sulfonated silica-carbon nanocomposites were remarkably
high, between 65 to 72% during the reaction course, and the
The time courses of styrene conversion and the selectivities
of DPB and MPI are shown in Figure 6. The reactivity of styrene
over these catalysts follows the order of Nafion NR-50>Si66C33-
673-SO3H>Si66C33-823-SO3H>Si50C50-673-SO3H>CBV720>
Si50C50-823-SO3H>Si33C66-673-SO3H. In contrast, S33C66-823-
SO3H, Sugar-673-SO3H, and Sugar-823-SO3H show almost no
conversion even after 24 h, indicating the diffusion limitations
of styrene molecules to sulfonic acid groups in these non-
porous structures. These internal sulfonic acid sites in Si33C66-
823-SO3H have been demonstrated not accessible with TMPO
molecules, see Figure 5A2. Considering the larger dynamic di-
ameter of styrene compared to TMPO, it is reasonable to
assume that styrene cannot access the carbon sheets of the
carbon phase in the composite Si33C66-823-SO3H. It is similarly
understandable that Si33C66-673-SO3H shows a rather low reac-
ChemCatChem 2015, 7, 3047 – 3058
www.chemcatchem.org
3054
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
The dimerization of a-methyl
styrene (AMS) was also per-
formed in the presence of these
sulfonated silica-carbon nano-
Table 2. Dimerization of styrene and AMS by sulfonated silica-carbon nanocomposite catalysts.
[SO3H][a]
Rate of
DPB[b]
[mmolhÀ1
TOF of
DPB[c]
[hÀ1
Rate of unsaturated
dimers from AMS[d]
[mmolhÀ1
TOF of unsaturated
dimers from AMS[e]
[hÀ1
]
[mmolgÀ1
]
]
]
]
composites.
The
reaction
Si33C66-673-SO3H
Si50C50-673-SO3H
Si66C33-673-SO3H
Si33C66-823-SO3H
Si50C50-823-SO3H
Si66C33-823-SO3H
CBV720[f]
0.57
0.40
0.31
0.37
0.25
0.15
0.56
0.80
15.1
153.3
436.2
trace
84.7
211.8
89.7
228.2
0.5
7.7
28.1
trace
4.6
28.2
3.2
412.2
935.2
1867.5
n.d.
n.d.
n.d.
6351.1
33.6
14.5
46.8
120.5
n.d.
n.d.
n.d.
Scheme of the dimerization of
AMS is shown in Scheme S2.
Three dimerization products, un-
saturated acyclic dimers (2,4-di-
phenyl-4-methyl-1-pentene (1-
226.8
0.8
DPMP)
and
2,4-diphenyl-4-
Nafion NR50
5.7
methyl-2-pentene (2-DPMP)) and
a saturated dimer (1,1,3-trimeth-
yl-3-phenylindan (TMPI)) were
detected. The time courses of
the conversion of AMS, the se-
lectivities of unsaturated dimers
and saturated TMPI are shown in
Figure 7. The selectivities for the
[a] The density of SO3H group was determined from titration. [b] The formation rate of DPB was calculated
from the early stage of dimerization of styrene. [c] TOF of DPB was calculated from the formation rate of DPB
normalized to per acid site. [d] The formation rate of unsaturated dimers was calculated from the early stage
of dimerization of AMS. [e] TOF of unsaturated dimers from AMS was calculated from the formation rate of un-
saturated dimers normalized to per acid site. [f] Acid sites in CBV720 was H+.
selectivities for MPI were all below 15%. It is largely due to the
moderate Brønsted acidity of sulfonic acid groups bound to
polycyclic aromatic carbon in these sulfonated silica-carbon
nanocomposites. Furthermore, the coexistence of phenolic OH
groups in thus pyrolyzed carbon may also account for the re-
striction of intramolecular Friedel–Crafts alkylation, as suggest-
ed recently by others.[45] Among all these catalysts tested, the
highest activity towards the formation of DPB was obtained
over Si66C33-673-SO3H (The selectivity of DPB reached 66% at
the conversion of 96% after 24 h), see the plots of the selectiv-
ity of DPB versus conversion of styrene in Figure S8.
unsaturated dimers (1-DPMP and 2-DPMP) are close to 100%
over the SimCn-673-SO3H catalysts, whereas over CBV720 and
Nafion NR50, side product TMPI is produced in a much larger
quantity. These results indicate that the acid strength SO3H
groups in the sulfonated silica-carbon nanocomposites can
protonate the C=C double bond of AMS for dimerization, but
not that strong for a Friedel-Crafts alkylation to the indane
dimer of AMS. The initial formation rates of unsaturated dimers
from AMS and the TOFs are listed in Table 2. Similar activity
order to the dimerization of styrene is observed for the dimeri-
zation of AMS over the SimCn-673-SO3H catalysts. With the in-
crease of silica-carbon ratio, the formation rate of unsaturated
dimers of AMS increases and the TOF increases more conspicu-
ously, showing the crucial effect of large mesoporosity for the
dimerization of AMS. The structural contribution of silica com-
ponent in these mesostructures ensures such accessible SO3H
groups for AMS molecules.
The formation rate of DPB, calculated from the early stage
of dimerization of styrene and the TOF (turnover frequency),
calculated from the formation rate normalized to per acid site,
are listed in Table 2. Of the silica-carbon nanocomposites pyro-
lyzed at the same temperature, the formation rate and TOF of
DPB both increase with the decrease of carbon content in the
nanocomposite. The nanocomposite of a lower carbon content
possesses a lower SO3H site density but shows a significantly
higher TOF of DPB, supporting that this reaction is controlled
by diffusion of styrene molecules to the SO3H sites in the
carbon domains in the nanocomposites. Seeing the physico-
chemical properties of sulfonated silica-carbon nanocompo-
sites in Table 1, entries 13–18, mesoporosity, rather than sur-
face area and total porosity, is found to nicely correlate with
the TOFs of DPB over these nanocomposites. For Si33C66-673-
SO3H with little mesoporosity, dimerization of styrene can only
slowly proceed. For Si50C50-673/823-SO3H and Si66C33-673/823-
SO3H where sulfonated carbon sheets are well incorporated in
the silica-supported pore walls with ꢀ7 nm mesopores open,
dimerization of styrene into DPB can be facilitated. Comparing
nanocomposites of the same starting silica-carbon ratio but
pyrolyzed at different temperatures, the TOFs of DPB are quite
comparable, due to the similar mesostructure and thus similar
spatial constraints for the dimerization of styrene. In addition,
the selectivities of MPI in the 823 K-pyrolyzed nanocomposites
are lower than their 673 K-pyrolyzed counterparts, probably
arising from the lower density of SO3H groups.
Conclusions
We have demonstrated that a broad family of silica-carbon ma-
terials can be synthesized by an EISA approach from the
system of sucrose-TEOS-F127 and the latter removal of carbon
or silica component, as shown in Scheme 3. By changing the
initial amount of sucrose, the pore structure of the silica-
carbon nanocomposites can be tuned while the silica part
keeps a 2D hexagonal mesoporous framework. A possible
mechanism of the formation of the silica-carbon composites
was proposed and analyzed with the combination of the re-
sults of SAXS, TEM, N2 sorption, and NMR. It is found that
silica-carbon composites after pyrolysis are mesoporous and
composed of homogeneously interpenetrating silica and
carbon components; while they lose mesoporosity when F127-
derived mesopores are occupied by the excessive carbonized
sucrose as in the case of Si33C66. Such series of silica-carbon
composites pyrolyzed at moderate temperatures were func-
tionalized with SO3H groups and can act as efficient and selec-
tive acid catalysts in the dimerization of styrene to DPB and di-
ChemCatChem 2015, 7, 3047 – 3058
www.chemcatchem.org
3055
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
merization of a-methylstyrene to unsaturated dimers. Acidity
was carefully analyzed with titration, FTIR and TMPO-probed
31P NMR to offer a clear understanding of these dimerization
reactions.
Experimental Section
Synthesis of ordered mesoporous silica-carbon composites
Mesoporous silica-carbon composites were synthesized by the
evaporation-induced triconstituent assembly method, wherein su-
crose was used as carbon precursor, prehydrolyzed TEOS as silica
precursor, and triblock copolymer F127 as template (Scheme 1).[35]
Three samples with different carbon content were synthesized. In
a typical synthesis, 6.4 g of block copolymer F127 (Pluronic F127,
EO106PO70EO106, Mw =12600) was dissolved in ethanol (32 g) with
concentrated HCl (37%, 0.3 g) and the mixture was treated in an
ultrasonic bath for 1 h to afford a clear solution. In parallel, sucrose
(5.7 g) was dissolved in deionized water (10 g). Then, the sucrose
solution and 8.32 g of TEOS were added into the mixture. The mix-
ture was sonicated for another 1 h to form a homogeneous solu-
tion. The solution was transferred into dishes to evaporate ethanol
at 313 K during 20 h following by heating treatment at 433 K for
24 h to thermopolymerize sucrose. The as-made product, named
as Si50C50, was scraped from the dishes. The composites with a dif-
ferent initial carbon content of 66 wt% and 33 wt%, denoted as
Si33C66 and Si66C33 respectively, were prepared according to a similar
procedure of Si50C50 by varying the mass ratios of sucrose to
TEOS. Pyrolysis was performed in a tubular furnace at 673 or 823 K
for 15 h or at 1073 K for 2 h under N2 flow to remove the F127
template and further carbonize sucrose. The pyrolyzed silica-
carbon composites were denoted as SimCn-T, where m and n repre-
sent the initial weight percentages of silica and carbon compo-
nents and T represents the pyrolysis temperature in Kelvin.
Synthesis of silica and carbon materials from silica-carbon
composites
The pyrolyzed nanocomposites, SimCn-1073, were calcined at 823 K
for 5 h in air to burn off carbon to obtain the pure mesoporous
silica samples, SimCn-Si. The nanocomposite was also immersed in
a 10 wt% HF solution (20 mL per gram of sample) at room temper-
ature for 24 h. The black solid was washed repeatedly with H2O
until FÀ was not detected with CaCl2. The resultant product was
named SimCn-C, corresponding to its parent nanocomposite.
&
*
Figure 7. Dimerization of AMS over ( ) Si33C66-673-SO3H, ( ) Si50C50-673-
~
^
SO3H, ( ) Si66C33-673-SO3H, (”) CBV720, and ( ) Nafion NR50. Reaction condi-
tions: AMS (0.73 g), catalyst (50 mg), toluene (5.22 g), reaction temperature
323 K.
Preparation of sulfonic acid group-functionalized meso-
porous silica-carbon composite
Nanocomposites pyrolyzed at moderate temperatures, SimCn-T (T
equals 673 or 823 K), were treated with concentrated sulfuric acid
(20 mL per gram of sample) in Teflon-lined autoclaves at 423 K for
15 h. The solid was washed extensively with hot distilled water
(>353 K) until no sulfate ions were detected with Ba(NO3)2. The
product was named SimCn-T-SO3H according to its parent nanocom-
posite.
Characterization
The SAXS patterns were recorded at the Dutch–Belgian beamline
(BM26) of the European Synchrotron Radiation Facility and
a custom software was used to reduce the measured 2D data into
Scheme 3. Summary of materials derived in this study by applying EISA syn-
thesis of F127, TEOS and sucrose. A, B, and C represent Si33C66, Si50C50, and
Si66C33, respectively.
ChemCatChem 2015, 7, 3047 – 3058
www.chemcatchem.org
3056
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
1D scattering patterns.[46] Average distance between neighboring
pores was calculated as 4p/ 3q*, where q* is the scattering vector
in the presence of the solid catalyst (50 mg) at 323 K. Samples
were withdrawn at intervals and analyzed by GC.
p
of the first peak. Nitrogen sorption isotherms were measured with
a Micromeritics Tristar 3000 apparatus at 77 K. Prior to measure-
ments, samples were pretreated at 393 K for 10 h in a N2 flow. The
Brunauer–Emmett–Teller surface area (SBET) was calculated over
a relative pressure range of 0.05 to 0.30. The total pore volume
(Vtot) was calculated from the saturation plateau of adsorption at
high relative pressures. Mesopore volume (Vmeso) was calculated
after subtraction of the micropore volume (Vmicro) as obtained by
the t-plot method. Average mesopore diameter (D) was calculated
from the adsorption branch of the isotherms, based on the Bruna-
uer–Joyner–Halenda (BJH) method. Transmission electron micros-
copy (TEM) images were acquired on a Philips CM30 microscope
operated at 300 kV. Samples were dispersed in ethanol by sonica-
tion and deposited onto a holey-carbon-film support grid. Ther-
mogravimetric analyses (TGA) were performed on a Q500 equip-
ment (TA Instruments, Brussels, Belgium). Weight losses were moni-
tored as the samples were heated from room temperature to
1073 K at 5 KminÀ1 under O2. 13C CP-MAS NMR spectra were col-
lected using a Bruker DSX400 spectrometer (B0 =9.4 T) at room
temperature with a frequency of 100.61 MHz, a recycle delay of
10 s and a contact time of 5.0 ms. The sample was packed in
a 4 mm zirconia rotor and the spinning frequency was 6 kHz. Tetra-
methylsilane was used as a chemical shift reference. The extremely
long relaxation delay of carbon materials made it infeasible to con-
duct single-pulse measurements for quantitative determinations.
Thus, the CP/MAS spectra of 13C were applied to qualitative com-
parisons among samples, assuming similar cross-polarization capa-
bilities of the 13C resonance signals. 29Si solid-state MAS NMR spec-
tra were collected using a Bruker AMX300 spectrometer (B0 =7.0 T)
at room temperature with a frequency of 59.63 MHz and a recycling
delay of 60 s. The sample was packed in a 4 mm zirconia rotor and
the spinning frequency was 5 kHz. Tetramethylsilane was also used
as a reference material. The Brønsted acid strength and content of
the sulfonated silica-carbon nanocomposites was examined by 31P
MAS NMR. The sample was first evacuated at 423 K for 2 h and im-
mersed in a tetrahydrofuran (THF) solution which contained an ad-
equate amount of trimethylphosphine oxide (TMPO) for 20 h. After
evacuation of removing the THF solvent, the sample was packed in
a 4 mm zirconia rotor in a N2 glovebox. 31P NMR spectra were col-
lected on a Bruker AMX300 spectrometer (B0 =7.0 T) with a recycle
delay of 60 s and a rotor spinning frequency of 6 kHz. An aqueous
solution of orthophosphoric acid (85%) was used as a chemical
shift reference.
Acknowledgements
The authors thank the long-term structural funding of the Flem-
ish government (Methusalem). They are also grateful to the Bel-
gian Science Policy (Belspo) for providing financial support (FdC)
and to the KU Leuven (OT 11/061). RZ thanks the China Scholar-
ship Council (No. 201206210307) financial support. We are grate-
ful to Kristof Houthoofd for his support in the NMR analysis.
Keywords: acid catalysis
heterogeneous catalysis
nanocomposite
·
dimerization of styrene
·
·
porous carbon silica-carbon
·
[1] F. de Clippel, M. Dusselier, S. Van de Vyver, L. Peng, P. A. Jacobs, B. F.
Sels, Green Chem. 2013, 15, 1398–1430.
[2] D. E. De Vos, M. Dams, B. F. Sels, P. A. Jacobs, Chem. Rev. 2002, 102,
3615–3640.
[3] A. Taguchi, F. Schuth, Microporous Mesoporous Mater. 2005, 77, 1–45.
[4] M. Hartmann, Chem. Mater. 2005, 17, 4577–4593.
[5] Z. Wu, D. Zhao, Chem. Commun. 2011, 47, 3332–3338.
[6] B. J. Scott, G. Wirnsberger, G. D. Stucky, Chem. Mater. 2001, 13, 3140–
3150.
[7] M. Manzano, M. Vallet-Regí, J. Mater. Chem. 2010, 20, 5593–5604.
[8] F. Schüth, Chem. Mater. 2001, 13, 3184–3195.
[9] S. Polarz, M. Antonietti, Chem. Commun. 2002, 2593–2604.
[10] Y. Wan, H. Yang, D. Zhao, Acc. Chem. Res. 2006, 39, 423–432.
[11] C. Liang, Z. Li, S. Dai, Angew. Chem. Int. Ed. 2008, 47, 3696–3717;
Angew. Chem. 2008, 120, 3754–3776.
[12] J. Shi, Z. Hua, L. Zhang, J. Mater. Chem. 2004, 14, 795–806.
[13] M. L. Anderson, R. M. Stroud, D. R. Rolison, Nano Lett. 2002, 2, 235–240.
[14] M. Choi, F. Kleitz, D. Liu, H. Y. Lee, W. S. Ahn, R. Ryoo, J. Am. Chem. Soc.
2005, 127, 1924–1932.
[15] F. de Clippel, A. Harkiolakis, X. Ke, T. Vosch, G. Van Tendeloo, G. Baron, P.
Jacobs, J. Denayer, B. Sels, Chem. Commun. 2010, 46, 928–930.
[16] F. de Clippel, A. Harkiolakis, T. Vosch, X. Ke, L. Giebeler, S. Oswald, K.
Houthoofd, J. Jammaer, G. Van Tendeloo, J. Martens, P. Jacobs, G. Baron,
B. Sels, J. Denayer, Microporous Mesoporous Mater. 2011, 144, 120–133.
[17] H. Kim, J. Cho, Nano Lett. 2008, 8, 3688–3691.
[18] T. Kwon, H. Nishihara, Y. Fukura, K. Inde, N. Setoyama, Y. Fukushima, T.
Kyotani, Microporous Mesoporous Mater. 2010, 132, 421–427.
[19] H. Nishihara, Y. Fukura, K. Inde, K. Tsuji, M. Takeuchi, T. Kyotani, Carbon
2008, 46, 48–53.
[20] K. Nakajima, M. Okamura, J. N. Kondo, K. Domen, T. Tatsumi, S. Hayashi,
M. Hara, Chem. Mater. 2009, 21, 186–193.
[21] Y. Wan, H. Wang, Q. Zhao, M. Klingstedt, O. Terasaki, D. Zhao, J. Am.
Chem. Soc. 2009, 131, 4541–4550.
Catalytic reaction
Acid catalysis of the sulfonated silica-carbon nanocomposites were
demonstrated by dimerization of styrene and a-methyl styrene
(AMS). For comparison, an ion-exchanged resin (Nafion NR50),
a zeolite (CBV720) and sulfonated sugar catalysts prepared from
direct pyrolysis of sucrose (Sugar-673-SO3H and Sugar-823-SO3H)[32]
were also tested under the same reaction conditions. Prior to reac-
tions, all catalysts were dehydrated at 423 K overnight. Dimeriza-
tion of styrene was performed in a closed glass vial. Styrene
(0.50 g) was reacted over the catalyst (50 mg) in toluene (5.00 g,
solvent) at 373 K in the presence of n-decane (0.73 g, internal stan-
dard) under magnetic stirring. Samples were withdrawn at regular
intervals from the reaction mixtures and analyzed by a GC-FID in-
strument using a capillary column. Similarly dimerization of a-
methyl styrene was also conducted in a closed glass vial with AMS
(0.73 g) and n-decane (0.73 g, internal standard) in toluene (5.22 g)
[22] M. Tsionsky, G. Gun, V. Glezer, O. Lev, Anal. Chem. 1994, 66, 1747–1753.
[23] Y. Mastai, S. Polarz, M. Antonietti, Adv. Funct. Mater. 2002, 12, 197–202.
[24] J. Pang, V. T. John, D. A. Loy, Z. Yang, Y. Lu, Adv. Mater. 2005, 17, 704–
707.
[25] B. Toury, F. Babonneau, J. Eur. Ceram. Soc. 2005, 25, 265–270.
[26] Z. Yang, Y. Xia, R. Mokaya, J. Mater. Chem. 2006, 16, 3417–3425.
[27] P. Valle-Vigón, M. Sevilla, A. B. Fuertes, Microporous Mesoporous Mater.
2010, 134, 165–174.
[28] S. S. Park, Y. Jung, C. Xue, R. Che, D. Zhao, C. Ha, Chem. Mater. 2010, 22,
18–26.
[29] R. Liu, Y. Shi, Y. Wan, Y. Meng, F. Zhang, D. Gu, Z. Chen, B. Tu, D. Zhao, J.
Am. Chem. Soc. 2006, 128, 11652–11662.
[30] Q. Hu, R. Kou, J. Pang, T. L. Ward, M. Cai, Z. Yang, Y. Lu, J. Tang, Chem.
Commun. 2007, 601–603.
[31] Y. Meng, D. Gu, F. Zhang, Y. Shi, L. Cheng, D. Feng, Z. Wu, Z. Chen, Y.
Wan, A. Stein, D. Zhao, Chem. Mater. 2006, 18, 4447–4464.
ChemCatChem 2015, 7, 3047 – 3058
www.chemcatchem.org
3057
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
[32] M. Toda, A. Takagaki, M. Okamura, J. N. Kondo, S. Hayashi, K. Domen, M.
Hara, Nature 2005, 438, 178–178.
[41] R. J. White, V. Budarin, R. Luque, J. H. Clarka, D. J. Macquarrie, Chem.
Soc. Rev. 2009, 38, 3401–3418.
[33] L. Peng, A. Philippaerts, X. Ke, J. Van Noyen, F. De Clippel, G. Van Tende-
loo, P. A. Jacobs, B. F. Sels, Catal. Today 2010, 150, 140–146.
[34] M. Kruk, M. Jaroniec, R. Ryoo, S. H. Joo, J. Phys. Chem. B 2000, 104,
7960–7968.
[35] S. Van de Vyver, L. Peng, J. Geboers, H. Schepers, F. De Clippel, C.
Gommes, B. Goderis, P. A. Jacobs, B. F. Sels, Green Chem. 2010, 12,
1560–1563.
[36] S. Schlienger, A. Graff, A. Celzard, J. Parmentier, Green Chem. 2012, 14,
313–316.
[37] Y. Wei, J. Xu, H. Dong, J. H. Dong, K. Qiu, S. A. Jansen-Varnum, Chem.
Mater. 1999, 11, 2023–2029.
[42] P. Valle-Vigón, M. Sevilla, A. B. Fuertes, Appl. Surf. Sci. 2012, 261, 574–
583.
[43] F. de Clippel, M. Dusselier, R. Van Rompaey, P. Vanelderen, J. Dijkmans, E.
Makshina, L. Giebeler, S. Oswald, G. V. Baron, J. F. M. Denayer, P. P. Pes-
carmona, P. A. Jacobs, B. F. Sels, J. Am. Chem. Soc. 2012, 134, 10089–
10101.
[44] A. Corma, H. Garcia, J. Chem. Soc. Dalton Trans. 2000, 1381–1394.
[45] S. Suganuma, K. Nakajima, M. Kitano, H. Kato, A. Tamura, H. Kondo, S.
Yanagawa, S. Hayashi, M. Hara, Microporous Mesoporous Mater. 2011,
143, 443–450.
[46] C. J. Gommes, B. Goderis, J. Appl. Crystallogr. 2010, 43, 352–355.
[38] A reference sample derived from the co-assembly of F127 and sucrose
without the addition of silica source was prepared and its N2 sorption
analysis results show no mesoporosity.
[39] W. Chu, S. Chiang, J. Li, S. Kuo, RSC Adv. 2014, 4, 784–793.
[40] M. Okamura, A. Takagaki, M. Toda, J. N. Kondo, K. Domen, T. Tatsumi, M.
Hara, S. Hayashi, Chem. Mater. 2006, 18, 3039–3045.
Received: June 25, 2015
Published online on September 3, 2015
ChemCatChem 2015, 7, 3047 – 3058
www.chemcatchem.org
3058
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim