Synthesis of carbon monolith with bimodal meso/macroscopic pore structure and its application in asymmetric catalysis

Article abstract of DOI:10.1016/j.cattod.2009.09.005

Carbon monolith with bimodal meso/macroporous pore structures was synthesized by the nanocasting method, using corresponding monolith silica as templates. The pore system of this material was characterized by instrumental analysis. The adjacent macropores are interconnected through uniform-sized windows, and the walls of macrospheres consist of mesostructured pores. Chiral Co(III)-(BF₃) salen immobilized on a meso/macroporous carbon monolith can be applied as an effective heterogeneous catalyst for an asymmetric reaction such as a hydrolytic kinetic reaction of terminal epoxides. The catalysts were prepared with different loading amounts of aluminium chloride to anchor the active salen complex, and the catalytic activity increased with increasing amount of aluminum chloride on the surfaces, showing up to 99 ee% of the product epichlorohydrin (ECH).

Full text of DOI:10.1016/j.cattod.2009.09.005

Catalysis Today 150 (2010) 91–99
Contents lists available at ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Synthesis of carbon monolith with bimodal meso/macroscopic pore structure
and its application in asymmetric catalysis
*
Yong-Suk Kim, Xiao-Feng Guo, Geon-Joong Kim
Department of Chemical Engineering, Inha University, 253 Younghyun-dong, Incheon, Nam-gu 402-751, Democratic People’s Republic of Korea
A R T I C L E I N F O
A B S T R A C T
Article history:
Carbon monolith with bimodal meso/macroporous pore structures was synthesized by the nanocasting
method, using corresponding monolith silica as templates. The pore system of this material was
characterized by instrumental analysis. The adjacent macropores are interconnected through uniform-
sized windows, and the walls of macrospheres consist of mesostructured pores. Chiral Co(III)-(BF3) salen
immobilized on a meso/macroporous carbon monolith can be applied as an effective heterogeneous
catalyst for an asymmetric reaction such as a hydrolytic kinetic reaction of terminal epoxides. The
catalysts were prepared with different loading amounts of aluminium chloride to anchor the active salen
complex, and the catalytic activity increased with increasing amount of aluminum chloride on the
surfaces, showing up to 99 ee% of the product epichlorohydrin (ECH).
Available online 17 October 2009
Keywords:
Bimodal pores
Mesoporous carbon monolith
SBA-15
Enantioselective catalysis
ß 2009 Elsevier B.V. All rights reserved.
1. Introduction
carbon materials directly into a monolithic shape. Recently, several
groups have reported the successful synthesis of monolithic carbons
For a variety of applications, it would be advantageous to have
material with pores in two or three different length scales in an
ordered fashion, with a hierarchical structure and interconnectiv-
ity between the pores. For catalytic applications, reactant
molecules need to readily access the interior pore structure.
Hierarchical materials containing both interconnected macropor-
ous and mesoporous structures have enhanced properties
compared with single-sized pore materials, due to increased mass
transport through the pore channels with a high specific surface
area at the molecular level.
Porous carbons fabricated by the nanocasting method, using
inorganic silicas as a mold, have attracted wide interest during the
past decade [1–11]. These carbons, characterized by high specific
surface area, uniform textural pores, and large pore volume, have
been applied in many areas including catalysis, separation, and
adsorption and energy storage [12,13]. Most carbons are generally
by nanocasting different templates, including silica monoliths [14–
18], particle arrays [19,20], and polymer foams [21,22]. By selecting
the proper templates, the morphology and pore structure of the final
carbon monoliths can be easily controlled to cater to various
applications. To date, two kinds of carbon monoliths, with different
porous structures, havebeenfabricated. Zhaoand Mokaya etal. have
utilized ordered mesoporous silica aggregates, via chemical vapor
deposition or polymerization followed by carbonization, to prepare
carbonmonoliths [15,16]. Theas-obtained carbonmonolithsfeature
a high surface area and exclusively mesoporous structure, but in
most cases, these materials do not have much practical value
because the interior of the carbons is not easily accessible [21,22].
Recently, some groups have reported carbon monoliths with a
hierarchical macro/mesoporous structure [17,18,21,22]. The pre-
sence of large macropores (also through-pores or perfusion-pores)
facilitates the rapid transport of species to the interior of monoliths,
whereas mesopores guarantee a large surface area for interaction.
The well-arranged, open porous structure of this kind of carbon
monolith renders it very useful for a variety of purposes. Further,
different macroscopic morphologies are needed for advanced
property research. Meso/macroporous carbon monolith with a
highly ordered pore size is an especially ideal material for an
electrochemical doubly layered capacitor (EDLC) as well as catalysis.
But so far, high-ordered meso/macroporous carbon monoliths with
three-dimensional (3D) structure using meso/macroporous silica as
a hard template, which would be ideal for these purposes, have not
been reported.
produced as a powder form in the size range from 1 to 10
mm.
However, under certain circumstances, the powder-like morphol-
ogy makes it difficult to use the material as is. For example, as an
electrode of a double layer capacitor, carbon powders must be glued
to form a flat layer. When used as a separation or adsorption media,
the fine carbon powders packed into a column may bring high back-
pressure.Onepossiblewaytosolvetheseproblemsistofabricatethe
* Corresponding author. Fax: +82 32 872 4046.
E-mail address: kimgj@inha.ac.kr (G.-J. Kim).
0920-5861/$ – see front matter ß 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.cattod.2009.09.005

92
Y.-S. Kim et al. / Catalysis Today 150 (2010) 91–99
In the present work, we report a new carbon monolith with
393 K. The meso/macroporous carbon monolith will be denoted as
C-SBA-15 and C-SBA-16, respectively.
bimodal meso/macroporous 3D structure. A simple and versatile
procedure, by coupling nanocasting and phase separation, was
proposed for its synthesis. This carbon monolith, having novel pore
structure with macroporous skeletons and mesoporous walls, was
employed as an efficient support to immobilize active salen
complexes for asymmetric catalysis. In this work, trends in activity
and enantioselectivity of chiral Co(III)-BF3 salen immobilized on
meso/macroporous carbon composites have been examined in a
hydrolytic kinetic resolution (HKR) of epoxide. These catalysts
exhibited very high activity and selectivity to form an optically
pure epoxide, with up to 99 ee%.
2.1.3. Surface oxidation of meso/macroporous carbon monolith
As-synthesized meso/macroporous carbon monolith (C-SBA-15
or C-SBA-16; 0.2 g) and 100 mL methylene chloride (MC) were
added to a 250 mL flask, and the suspension was vibrated
ultrasonically for 0.5 h. After addition of phase transfer agent
tetrabutylammonium bromide (TBABr, TCI Co.; 1.0 g) dissolved in
the mixed solution of H2O (10 mL) and acetic acid (10 mL), as well
as an aqueous solution of potassium permanganate (5.0 g;
dissolved in 5 mL H2O), the mixture was stirred vigorously at
298 K for 24 h. The powder was washed with aqueous HCl and
methanol. The washing with methanol was repeated at least four
times. The sample was dried in a vacuum oven overnight at 338 K.
2. Experimental
2.1. Synthesis of meso/macroporous silica template and
mesoporous carbon monolith
2.1.4. Immobilization of chiral salen catalyst on the bimodal
porous materials
The anhydrous aluminum chloride was immobilized on the
ordered mesoporous carbon monolith in order to anchor the chiral
salen complex as an active site for the enantioselective catalyst.
Aluminum chloride was dissolved in dried tetrahydrofuran (THF),
and the meso/macroporous carbon monolith having the oxidized
surfaces (or silica SBA-16) was added to that solution. The loading
amount of aluminum chloride was controlled from 10 wt% to
30 wt% (Al basis) to the functionalized meso/macroporous carbon
monolith support. After evaporation of solvent, the recovered solid
sample was calcined at 823 K for 2 h under nitrogen.
The chiral Co(III)-(BF3) salen was prepared by the same method
as shown in the previous paper [23]. The structure of chiral Co(III)-
(BF3) salen is indicated in Scheme 1. The Al-containing mesoporous
C-SBA-16 or SBA-16 silica was used as a support for immobilization
of chiral (salen) cobalt complexes. The procedure to anchor the
chiral complexes on the surfaces of meso/macroporous C-SBA-16 is
also shown in Scheme 1. The heterogenized chiral salen catalysts
were prepared by refluxing Al-containing mesoporous C-SBA-16
monolith with the solution containing the chiral Co(III)-(BF3) salen
complex in MC for 2 h. The sample, with a dark green color, was
obtained by filtration and sequential washing with MC, THF and
methanol until the filtrate was colorless. The catalyst was dried in
vacuo to yield a heterogenized Co (salen) complex.
2.1.1. Synthesis of meso/macroporous SBA-15 or -16 silica monolith
The meso/macroporous composites were synthesized from the
mixture of two types of prehydrolyzed silica sol (SBA-15 and SBA-
16) and PMMA polymer spheres. The typical synthetic procedure
for the mesoporous SBA-15 sol is as follows: first, the calculated
amount of poly(ethylene oxide)-block-poly(propylene oxide)-
block-(ethylene oxide) (EO20 PO70 EO20) (P-123); numbers in
subscripts denote the molar ratios of each component) was
dissolved in EtOH to make 35 wt% solution. To prepare the
prehydrolyzed precursor solution, tetraethyl orthosilicate (TEOS)
was dissolved in EtOH solution acidified by aqueous hydrochloric
acid (HCl). After mixing the two solutions, the mixture was
refluxed for 8 h, and then the solvent was partially evaporated. The
initial molar ratio of TEOS:EtOH:H2O:HCl was 1:3:4:0.04.
In a typical preparation of SBA-16 sol, 2.0 g of (ethylene
oxide)106–(propylene oxide)70–(ethylene oxide)106 (Pluronic F-
127) was dissolved in the solution of EtOH(4.69 g), H2O(0.94 g) and
2M-HCl(0.15 g) while stirring at 298 K. Into that solution, 6.25 g of
TEOS was added slowly while stirring at 308 K for 1 h. The mixture
was aged at 353 K for 2 h without stirring to evaporate the solvent,
until the viscous solution was recovered. Monodispersed PMMA
spheres with a diameter of 30
mm (TAFTICTM FH-S series; for use
with a light diffuser) were purchased from Toyobo Co. The SBA-15
or SBA-16 silica sol solution made by sol–gel process was added to
the same volume of PMMA spheres. The whole mixture was stirred
at room temperature, and then centrifuged in a polypropylene tube
with 13,000 rpm. Then the mixture was dried slowly at room
temperature for 5 days. The PMMA spheres were removed by
direct calcination at 873 K for 4 h in air. The heating rate was
maintained at 1 K/min.
2.1.5. General procedure for the catalytic reaction
In a representative reaction between epichlorohydrin (2-
(chloromethyl)oxirane; ECH) and water, the catalyst (1.0 mol%
salen-loading relative to ECH/support), THF solvent (5 mL) and (Æ)-
ECH (0.93 g, 10 mmol) were charged in an oven-dried 25 mL flask and
the reaction mixture was stirred in open atmosphere at ambient
temperature. After addition of water (81 mg, 4.5 mmol), the resultant
mixture was stirred for 40 h, and enantiomeric excess percentage
(ee%) values of the ring-opened product were determined by GC using
a chiral capillary column (CHIRALDEX, g-cyclodextrin trifluoroacetyl,
20 m  0.25 mm i.d.), and by HPLC using a Chiralcel1 OD-H column
(24 cm  0.46 cm).
2.1.2. Synthesis of meso/macroporous carbon monolith
The anhydrous 1.0 M-ZnCl2 solution in ethylether was incor-
porated into the pores of calcined meso/macroporous silica
monolith for two times. ZnCl2 was impregnated on the surfaces
of silica after evaporation of ether to generate the catalytic sites for
the polymerization of fulfuryl alcohol. The polymerization was
carried out by heating the ZnCl2-impregnated silica in fulfuryl
alcohol at 333 K for 3 h. The resulting solid sample was dried at
398 K for 12 h after washing the outer surfaces with ethanol. The
silica/resin composites were heated under a nitrogen atmosphere
at the heating rate of 5 K/min to 1173 K, and held at this
temperature for 7 h to carbonize the polymer inside the mesopores
of SBA type silica. The carbon–silica composite obtained after
pyrolysis was treated with 5 wt% hydrofluoric acid at room
temperature to remove the silica template. The template-free
carbon product was filtered, washed with ethanol, and dried at
2.1.6. Characterization
X-ray powder diffraction (XRD) data were acquired on a D/MAX
2500 V/PC diffractometer using Cu K
a radiation. The data were
collected from 0.6 to 38 (2 ) with a resolution of 0.028. The
u
morphology and microstructures of as-prepared samples were
characterized by field emission transmission electron microscopy
operating at 200 kV (FE-TEM, S-4200). The morphology and size of
macropores were determined by scanning electron microscopy
(SEM) with a Hitachi S-4200 microscope. SEM-EDX was used to
analyze the anchored cobalt contents. Nitrogen adsorption and
desorption isotherms were measured at 77 K using a micromeritics

Y.-S. Kim et al. / Catalysis Today 150 (2010) 91–99
93
Scheme 1. Schematic illustration of Co (III) salen complexes immobilized onto the functionalized meso/marcoporous carbon monolith.
ASAP 2010 system. The data were analyzed using the Brunauer–
Emmett–Teller (BET) model. Both adsorption and desorption
branches were used to calculate the pore size distribution by
the Barrett–Joyner–Halenda method. The pore volumes were taken
at the P/P0 = 0.995 single point. The specific surface areas were
calculated according to BET theory. The dispersivity of carbon
monolith and functionalized carbon monolith in methanol solvent
was tested by a Turbiscan analyzer (Fomulation Turbiscan Lab).
Fig. 1. Photographs of the silica SBA-15 (A); carbon monolith C-SBA-15 (B); silica monolith SBA-16 (C); carbon C-SBA-16 (D).

94
Y.-S. Kim et al. / Catalysis Today 150 (2010) 91–99
The coordination of Al on the C-SBA-16 and silica SBA-16 was
characterized by 27Al-MAS NMR spectroscopy (DSX 400 MHz Solid
State Bruker NMR(at KBSI Daegu); 400 MHz with magic angle
shown in Fig. 1. The silica sol became a white solid monolith having
the same shape as the molding container, whereas mesoporous C-
SBA-15 and C-SBA-16 monolith are black in color, maintaining the
shape of the silica templates. However, large and moldable meso/
macroporous carbon monolith could be obtained successfully in
centimeter scale, maintaining the same shape of container.
Obvious shrinkage in volume was observed after carbonization.
Scanning electron microscope (SEM) images of as-synthesized
silica and carbon monoliths are shown in Fig. 2. As-synthesized
spinning at 10 kHz,
p/2 pulse, 2 ms contact time, a repetition delay
of 2 s and 30,000 scans).
3. Results and discussion
3.1. Characterization of meso/macroporous monoliths
silica monolith (SBA-15) shows a uniform macropore size of 30
mm
and a highly ordered 3D well-interconnected network through
macroporous windows. It can be seen that large fractions of the
calcined sample were highly ordered in three dimensions and the
remaining portions exhibited uniform macroporosity intercon-
nected through windows whose diameter is typically the same as
that of the PMMA template used. In C-SBA-15 sample, a 3D well-
interconnected network was formed with an average macropore
The centimeter-scale silica monoliths were synthesized by
using the corresponding SBA-15 or SBA-16 sol and PMMA template
to control the bulk shape and macropore structure. The particle
size distribution of PMMA was very narrow, and the particles were
spherical in shape with a uniform size of around 30
mm. In this
case, the silica sol filled the void volume between the cross-packed
PMMA spheres. The SBA-15 or SBA-16 silica sol was solidified after
removal of solvents to form the monolith structure (Fig. 1). The
shape of the meso/macroporous silica monolith appears to
replicate the centrifuge tube shape, as shown in Fig. 1. In addition,
the size of the macroporous channels could be controlled by
changing the size of the template polymer spheres. During the
calcinations to remove the PMMA templates, densification of the
silica network occurred, resulting in shrinkage of the monolith.
Nanocasting by using two- or three-dimensional (2D or 3D)
structured mesoporous silica as a mold was applied to fabricate the
C-SBA-15 or C-SBA-16 monolith. The apparent morphologies of
those meso/macroporous carbon monoliths were replicated by the
corresponding meso/macroporous silica monolith template, as
channel size of 30
mm (Fig. 2B). As-synthesized silica monolith
(SBA-16) also shows the highly ordered macropore channels
interconnected 3-dimensionally through windows, similar to the
structure of silica monolith (SBA-15). It was observed that carbon
monolith C-SBA-16 has a much thicker wall than C-SBA-15.
Evidence of ordered mesopores in the meso/macroporous silica
monolith was found by XRD, TEM and N2 adsorption/desorption
analysis. For studying the interior structures and the morphology
of silica and carbon monoliths, samples were further analyzed by
TEM. Silica monolith SBA-15 and -16 samples prepared in this
study had the ordered mesoporous pore inside the macroporous
walls (Fig. 3A and C). The C-SBA-15 and C-SBA-16 monolith
Fig. 2. SEM images of the silica monolith SBA-15 (A); carbon monolith C-SBA-15 (B); silica monolith SBA-16 (C); C-SBA-16 (D).

Y.-S. Kim et al. / Catalysis Today 150 (2010) 91–99
95
Fig. 3. TEM images of the silica SBA-15 (A); carbon monolith C-SBA-15 (B); silica SBA-16 (C); and carbon monolith C-SBA-16 (D).
showed partly disordered mesopores distributed inside the wall of
the macropores (Fig. 3B and D). Silica monolith SBA-16 as well as C-
SBA-16 showed mesoporosity throughout the whole of the SBA-16
matrix. As can be seen in Fig. 3, the overlapped void volume of
spherical macropores and also the mesopores in the silica or
carbon walls are visible in TEM images. An array of interconnected
macropore structures over a large area provides more openings of
porous channels.
Fig. 4 shows the X-ray diffraction patterns of silica SBA-15, SBA-
16 made by the sol–gel process, and carbon monoliths C-SBA-15
and C-SBA-16 fabricated by the nanocasting method. In Fig. 4, SBA-
15 silica shows a very intense (1 0 0) diffraction peak with
additional three weak (1 1 0, 2 0 0, and 2 1 0) peaks after
calcination, indicating the well-developed 2D hexagonal pore
structure. The meso/macroporous carbon monolith C-SBA-15
showed a similar XRD pattern to that of silica SBA-15, but the
peak for (1 0 0) diffraction was shifted higher. This means that the
pore size of carbon monolith C-SBA-15 was decreased as compared
to that of its parent template SBA-15 silica. For determination of
the mesopore structure and the symmetry of SBA-16, XRD analysis
was performed, and the diffraction patterns of SBA-16 are shown in
Fig. 4. The reflection patterns could be indexed to the structure of
the mesopores. SBA-16 exhibited an intense (1 0 0) diffraction
peak with a distinctive shoulder peak due to the (0 0 2) plane. In
addition, the broad and unresolved peak for the (1 0 1) plane at 1.58
(2u) from the sample of SBA-16 silica obtained from the sol was
investigated, revealing the 3D hexagonal symmetry of its pore
structures. In Fig. 4D, XRD patterns of the meso/macroporous
Fig. 4. X-ray diffractograms of silica monolith SBA-15, -16, carbon monolith C-SBA-
15 and C-SBA-16.

96
Y.-S. Kim et al. / Catalysis Today 150 (2010) 91–99
Table 1
Pore structural parameters of meso/macroporous C-SBA-15 and -16 carbon monolite, as well as the parent silica template SBA-15 and -16.
Specific surface area (m2/g)
Pore volume (cm3/g)
Averaged pore size (A)
˚
Entry
Type of support
1
2
3
4
5
6
Silica SBA-15
721
650
851
901
660
666
0.92
0.98
1.17
1.24
0.66
0.62
78.3
62.0
66.0
75.6
54.6
48.0
Silica SBA-16
Carbon SBA-15
Carbon SBA-16
10 wt% Al/carbon SBA-15
30 wt% Al/carbon SBA-15
carbon monolith C-SBA-16 are also listed, and this carbon replica
obtained from SBA-16 silica monolith exhibited a diffraction
pattern similar to the SBA-16 silica, showing the same pore
structures.
The pore structural parameters of the parent SBA silica and their
carbon replica materials are shown in Table 1. Large surface area,
pore volumes, and pore sizes were formed in the meso/macro-
porous carbon composites. As can be seen in Table 1, it is clear that
the mesoporosity of the carbon material was maintained after
loading of 20 wt% AlCl3 and calcination at 873 K in nitrogen, but it
was partially blocked by Al compounds in the mesopore channels.
The surface area, pore volume, and mean pore size were decreased
after impregnation of aluminum sources.
The nitrogen physisorption isotherms for four types of monolith
samples are plotted in Fig. 5. The isotherm of silica monolith SBA-
15 (Fig. 5A) is of type IV according to IUPAC classification [24],
having a typical H1 hysteresis loop for materials with hexagonal
pore channels. The N2 adsorption–desorption isotherm of calcined
C-SBA-15 displays steep changes with increasing pressure, which
is an indication of the presence of mesoporosity (Fig. 5C). In Fig. 5B,
silica monolith SBA-16 indicated an ordered cage-like mesopore,
showing the typical H2 type hysteresis. However, carbon monolith
C-SBA-16 showed a combined hysteresis curve of H4 and H2.
To investigate the degree of functionalization of hydroxy
groups, dispersion of parent carbon monolith and surface-oxidized
carbon monolith C-SBA-15 was analyzed by Turbiscan using
different solvents (water, methanol, and toluene). For this analysis,
the monolith sample was ground into fine powders. The parent
carbon monolith was aggregated, and settled down completely in
water and toluene solvent. They exhibited good dispersal stability
in methanol even after several weeks, indicating strong hydro-
philicity of their surfaces. However, as-synthesized monolith
aggregated completely even using methanol as solvent. Quanti-
tative analysis using a novel optical analyzer (Turbiscan) was also
employed to measure the dispersion stability of hydroxy-
functionalized carbon monolith in water, toluene, and methanol.
The interpretation of the transmission profiles was adopted to
determine the dispersivity in the solvent by comparing the change
in light transmission caused by the sedimentation of the carbon
monolith in a sample cell. When sedimentation takes place in the
suspension, the transmission profiles vary with the dispersed
height of the sample with time. On the other hand, if the dispersion
is stable, no discernible change in the transmission profile with
time is observed over the entire sample cell. The aggregation and
sedimentation behavior was monitored for each carbon sample,
and the results of the transmission profile obtained in 15 min
intervals are plotted as the transmission intensity versus time in
Fig. 6.
3.2. Immobilization of chiral salen complexes
In this work, the homogeneous chiral Co(III)-BF3 salen catalyst
was chosen as a candidate to immobilize on the surfaces of Al-
containing mesoporous C-SBA-15 or -16 monolith. UV, FT-IR, ESCA,
Fig. 5. N2 adsorption–desorption isotherms of silica SBA-15 (A); silica SBA-16 (B); carbon monolith C-SBA-15 (C); and C-SBA-16 (D).

Y.-S. Kim et al. / Catalysis Today 150 (2010) 91–99
97
Fig. 6. Dispersion of parent carbon monolith and hydroxy-functionalized carbon monolith.
and NMR analyses were performed to determine the structure of
the synthesized homogeneous cobalt-salen catalysts, and the
results were reported previously [23]. These Co (salen) complexes
were easily attached onto the acidic sites of supports, and they
adhered strongly to the solid surfaces even after washing with
various polar solvents. We believe this unique immobilization is
due to the adsorption of the complex on to Lewis acid sites of
aluminum present on the support, since the homogeneous Co(III)-
BF3 salen was also attached strongly to the surfaces of pure Al2O3.
However, the Co(III)-salen having BF3 group was easily attached
the different loading amounts. There are two distinct peaks on the
27Al-MAS-NMR spectra of 30 wt% AlCl3-loaded carbon (Fig. 7A).
The strong peak at 0 ppm can be attributed to the octahedral
coordination (Lewis acidic sites), and the chemical shift of 60 ppm
is related to the aluminum atoms with tetrahedral coordination
¨
(BrOnsted acidic sites). For the carbon sample containing 10 wt%
AlCl3 (Fig. 7B), only the Lewis acidic Al-sites appeared without the
¨
presence of BrOnsted acid sites. When 30 wt% of AlCl3 was
impregnated on the SBA-16 silica, three distinct peaks on the 27Al-
MAS-NMR spectra were found as in Fig. 7C. The peak at chemical
shift of 20 ppm in the spectrum is interpreted as 5-coordinated Al-
site: namely these aluminums are surrounded by 4 oxygens in the
framework with one adsorbed molecule such as water at the non-
framework position [25]. There are many vacant tetrahedral sites
and OH silanol groups on the surfaces of SBA-16 silica. In general,
the silanol nest present on the surfaces of SBA-16 can be combined
with aluminum species to form the tetrahedral sites or octahedral
sites. The 27Al-MAS-NMR spectra clearly showed that Lewis acidic
sites were formed mainly on the surfaces of mesoporous carbon
¨
onto the hydronium ions such as BrOnsted acidic solids as reported
¨
in the paper. For instance, SiO2–Al2O3 support having BrOnsted
acid sites mainly could immobilize the chiral Co(III)-BF3 salen in a
large amount.
To perform the immobilization of chiral Co(III)-BF3 salen on the
surfaces of carbons, anhydrous AlCl3 was impregnated first on the
carbons or SBA silica in the range of 10–30 wt%, and the dried
sample was calcined at 550 8C in N2 stream. In present work, 27Al-
MAS-NMR analysis was performed to distinguish the coordina-
tion state of aluminium compound which were formed after Al
loading on the supports. Fig. 7 shows the 27Al-MAS-NMR spectra
of 30 wt% Al-incorporated SBA-16 silica and carbon samples with
¨
by attachment of AlCl3. Higher concentration of BrOnsted acidic
sites was formed on the surfaces of SBA-16 silica than mesoporous
Fig. 8. FT-IR spectra of C-SBA-16 support (A); chiral salen complex immobilized on
10 wt% AlCl3 loading C-SBA-16 (B); chiral salen complex immobilized on 20 wt%
AlCl3 loading C-SBA-16 (C); chiral salen immobilized on 30 wt% AlCl3 loading C-
SBA-16 (D); and pure chiral homogeneous salen complex (E).
Fig. 7. 27Al-MAS-NMR spectra of 30 wt% AlCl3-loaded C-SBA-16(A); 10% AlCl3-
loaded C-SBA-16 (B); 30 wt% AlCl3-loaded SBA-16 silica (C) (samples were calcined
at 550 8C in N2).

98
Y.-S. Kim et al. / Catalysis Today 150 (2010) 91–99
¨
carbons. However, the BrOnsted acid site itself may result in the
fatal drawback of racemization during the asymmetric reactions.
The pure mesoporous carbon, homogeneous chiral Co (III) salen
cobalt complexes and those immobilized on the carbon support
were characterized by FT-IR analysis (Fig. 8). The mesoporous SBA-
15, -16 and the replica carbons can supply the vacant spaces
enough to introduce the big salen complexes inside the pore
channels. The characteristic peaks for homogeneous chiral cobalt
(III) salen complex appeared at the position of 2958–2950, 2912,
1612 and 1535 cmÀ1, respectively. Those peaks were not found on
the IR spectrum of carbon monolith support (Fig. 8A), but the chiral
cobalt (III) salen complex immobilized on the meso/macroporous
carbon support (Fig. 8B–D) exhibited the similar absorption IR
bands as the pure homogeneous cobalt (III) salen complex (Fig. 8E),
showing that the salen complex was successfully attached on the
surfaces of mesoporous carbon monolith.
same conditions. The enantioselective ring-opening reaction of
epoxides was found to be applicable for the synthesis of an
enantiopure terminal epoxide using water as a nucleophile.
As can be seen in Fig. 9, higher reaction rates and ee% of product
ECH were observed when the high concentration of active sites
(salen complex) was allowed on the same amount of supports with
same reaction times. The anchored amount of salen complex was
proportional to that of Al present on the supports (EDX result in
Fig. 10). The attached contents of Co-salen complex were 4.1, 7.8
and 9.7 wt% over 10, 20 and 30 wt% Al-loaded mesoporous meso/
macroporous C-SBA-16, respectively. The remarkable correlation
between the reaction rates and the population of salen catalysts on
the support was evident from the result of catalysis. When the
aluminum chloride content was high, the action of the chiral
catalysts became superfluous. This result gives a strong support to
a cooperative bimetallic mechanism involving a simultaneous
activation of both epoxide and nucleophile by different cobalt-
salen units.
3.3. Asymmetric catalysis on heterogenized salen catalysts
The recyclability of catalysts was investigated in HKR of (Æ) ECH
by using different Al-loaded carbon catalysts. The catalyst was
collected by simple washing with MC, THF solvent for reuse after
completion of reaction. The catalytic activity and enantioselectivity
decreased slightly after fourth-time reuse, as shown in Fig. 11. When
the catalyst was used without further regeneration, it was
To observe the trends in the activity and enantioselectivity of
chiral cobalt-salen complexes immobilized on the meso/macro-
porous carbon C-SBA-16 and silica SBA-16, the reactivities of the
homogeneous and heterogenized chiral (salen) Co-BF3 complexes
were examined in the HKR of (Æ) ECH with a prolonged reaction
time. The reaction results are summarized in Fig. 9. The reaction with
catalyst 30 wt%-Al/C-SBA-16 led to 99 ee% of ECH after 28 h, and the
turn over frequency (TOF) at the initial reaction stage was 6.6 hÀ1
.
Whereas 30 wt%-Al/SBA-16 silica catalyst having the similar pore
structure gave a lower TOF of 5.0 hÀ1. The highest TOF of 7.1 hÀ1 was
shown for the homogeneous Co-salen catalyst. It is evident that the
immobilized Co(III)-BF3 salen catalyst on the meso/macroporous C-
SBA-16 exhibited high activity and selectivity as the homogeneous
one. However, with the chiral salen complex over SBA-16 meso/
macroporous silica, a lower catalytic activity was obtained even after
a prolonged reaction time. This result indicates that the character-
istics of the support surface played an important role in the catalysis.
The reactant water can be easily adsorbed on the hydrophilic surfaces
of silica SBA-16 (or -15) compared to the hydrophobic surfaces of
carbon C-SBA-16 (or -15) support, delaying the reactions under the
Fig. 9. Effect of loading amount of Al added to the support (SBA-16 carbon and silica
monolith) for immobilization of the salen complex on the catalytic activity in HKR
of racemate ECH (1.0 mol% salen relative to ECH was loaded in the system).
Fig. 10. Energy dispersive X-ray analysis pattern of 10 wt% AlCl3 loading C-SBA-16
(A); 20 wt% AlCl3 loading C-SBA-16 (B).

Y.-S. Kim et al. / Catalysis Today 150 (2010) 91–99
99
The adjacent macropores are interconnected through uniform-
sized windows, and the walls of these macrospheres consist of
mesostructured pores. On the basis of asymmetric HKR of ECH with
various catalysts, the chiral Co(III)-(BF3) salen immobilized on the
functionalized meso/macroporous carbon monoliths obtained in
the present study can be applied as effective heterogenized
catalysts for asymmetric reactions. The catalysts were prepared
with different loading amounts of aluminum chloride to anchor the
active salen complex, and the catalytic activity increased with
increasing amount of aluminum chloride on the surfaces. The
robust catalytic properties of meso/macroporous carbon materials
may expand the applications of porous materials. Furthermore,
such carbon materials with bimodal meso/macroscopic pores
would be useful for many potential applications, such as
heterogeneous catalysts when bulky or highly hydrophilic
molecules are involved as a reactant.
References
[1] R. Ryoo, S.H. Joo, S.J. Jun, J. Phys. Chem. B 103 (1999) 7743.
[2] R. Ryoo, S.H. Joo, M. Kruk, M. Jaroniec, Adv. Mater. 13 (2001) 677.
[3] S. Jun, S.H. Joo, R. Ryoo, M. Kruk, M. Jaroniec, Z. Liu, T. Ohsuna, O. Terasaki, J. Am.
Chem. Soc. 122 (2000) 10712.
[4] S.H. Joo, S.J. Choi, I. Oh, J. Kwak, Z. Liu, O. Terasaki, R. Ryoo, Nature 412 (2001) 169.
[5] S.S. Kim, T.J. Pinnavaia, Chem. Commun. (2001) 2418.
[6] S.B. Yoon, J.Y. Kim, J.S. Yu, Chem. Commun. (2001) 559.
[7] S.J. Han, T. Hyeon, Chem. Commun. (1999) 1955.
Fig. 11. Recyclability of catalysts in HKR of (Æ) ECH by using different amounts of Al-
loaded SBA-16 carbon and silica monolith catalysts (1.0 mol% salen relative to ECH was
loaded in the system. No further regeneration was applied for repeated use).
[8] S.J. Han, T. Hyeon, Carbon 37 (1999) 1645.
[9] J. Lee, K. Sohn, T. Hyeon, J. Am. Chem. Soc. 123 (2001) 5146.
[10] J. Lee, K. Sohn, T. Hyeon, Chem. Commun. (2002) 2674.
[11] J. Lee, J. Kim, T. Hyeon, Chem. Commun. (2003) 1138.
[12] H.C. Foley, Microporous Mater. 4 (1995) 407.
investigated that a little amount of leaching of homogeneous salen
from the support occurred. It is noteworthy that the heterogeneous
catalyst could be successfully recycled without much loss in activity
after re-attachment of chiral Co(III)-BF3 salen on the reused catalysts
in MC solvent. It was easy to isolate the immobilized salen catalysts
from the product solution containing epoxide and diol.
[13] T. Kyotani, Carbon 38 (2000) 269.
[14] B.H. Han, W.Z. Zhou, A. Sayari, J. Am. Chem. Soc. 125 (2003) 3444.
[15] H.F. Yang, Q.H. Shi, X.Y. Liu, S.H. Xie, D.C. Jiang, F.Q. Zhang, C.Z. Yu, B. Tu, D.Y. Zhao,
Chem. Commun. (2002) 2842.
[16] Y.D. Xia, R. Mokaya, J. Phys. Chem. C 111 (2007) 10035.
[17] A. Taguchi, J.H. Smatt, M. Linden, Adv. Mater. 15 (2003) 1209.
[18] Z.G. Shi, Y.Q. Feng, L. Xu, S.L. Da, M. Zhang, Carbon 41 (2003) 2677.
[19] O. Klepel, H. Straub, A. Garsuch, K. Bohme, Mater. Lett. 61 (2007) 2037.
[20] X.Q. Wang, K.N. Bozhilov, P.Y. Feng, Chem. Mater. 18 (2006) 6373.
[21] S. Alvarez, J. Esquena, C. Solans, A.B. Fuertes, Adv. Eng. Mater. 6 (2004) 897.
[22] S. Alvarez, A.B. Fuertes, Mater. Lett. 61 (2007) 2378.
[23] Y.S. Kim, C.-Y. Lee, G.-J. Kim, Bull. Korean Chem. Soc. 30 (2009) 1771.
[24] IUPAC reporting physisorption data for gas/solid systems with special reference
to the determination of surface area and porosity, Pure Appl. Chem. 57 (1985)
603.
4. Conclusions
The meso/macroporous materials having bimodal pore struc-
tures were synthesized, and their pore system was characterized
by instrumental analysis and by comparing the catalytic activity at
room temperature. Large carbon monoliths could be obtained in
centimeter scale from its corresponding template silica monolith.
[25] J. Gan, T. Wang, Z. Liu, W. Tan, Stud. Surf. Sci. Catal. 170B (2007) 1567.