Carbon silica composites for sulfur dioxide and ammonia adsorption

Amanda M.B. Furtado, Yu Wang, M. Douglas LeVan

Department of Chemical and Biomolecular Engineering, Vanderbilt University, Nashville, TN 37235, USA
a r t i c l e i n f o
Article history:
Received 6 April 2012
Received in revised form 14 July 2012
Accepted 15 July 2012
Available online 4 August 2012
Keywords:
Composite adsorbent
Carbon
Mesoporous silica
Sulfur dioxide
Ammonia
a b s t r a c t
This work focuses on creating nanoporous carbon silica composites from MCM-41 and two carbon
sources, sucrose and furfuryl alcohol. The carbon silica composite with sucrose as the carbon phase
was synthesized using a novel low temperature procedure. These novel, biphasic materials were tested
for their ability to adsorb two distinct types of gases: sulfur dioxide, an acidic gas, and ammonia, a basic
gas. The materials are characterized by XRD, nitrogen adsorption isotherms, high resolution TEM, and
TGA. The characterization techniques show that impregnation with the carbon phases does not disrupt
the hexagonal mesoporous silica structure. Equilibrium breakthrough results show that the presence
of the carbon phase enhances both ammonia and sulfur dioxide adsorption capacities compared to the
parent MCM-41 and BPL activated carbon.
2012 Elsevier Inc. All rights reserved.
1. Introduction
The carbon silica composite (CSC) introduced by Glover et al. [1]
in 2008 consists of carbonized polyfurfuryl alcohol within the
pores of MCM-41. This was the rst in a series of studies reported
by our group to produce biphasic composite materials targeting
light gas adsorption. The CSC provides a large surface area and
two distinct phases in which adsorption can occur: a nonpolar car-
bonaceous phase and a polar siliceous phase. The CSC material is of
interest since it is well documented that interactions between the
adsorbate and adsorbent are affected by the polarity of each; non-
polar surfaces such as carbon show greater attraction for adsorbate
molecules of low polarity, whereas polar surfaces have higher
afnity for polar molecules [2]. Materials that provide both polar
and nonpolar surfaces for adsorption have an advantage over
single-phase adsorbents in broad scale applications.
In todays society, light gases are widely used as industrial
chemicals, yet they can also be hazardous to human health [3].
Adsorbent materials designed for air purication applications must
have the ability to remove low concentrations of a broad spectrum
of toxic gases. For air purication in industrial settings, activated
carbons such as BPL AC are generally used [4]. Such carbons have
a large pore size distribution that provides macropores and
mesopores to enhance the transport properties throughout the
adsorbents and micropores that provide capacity for physical
adsorption due to strong potential wells [5]. The unique transport
properties and physical characteristics of biphasic materials have
been studied extensively [68]. Biphasic materials such as carbon
silica composites have the potential to be more efcient at air puri-
cation than single phase adsorbents such as activated carbon.
Each phase in the biphasic composite can be tailored to remove
one type of gas; i.e., silicas can be used to target the removal of
basic gases such as ammonia and an organic phase can be used
to target the removal of acidic gases such as sulfur dioxide.
Mesoporous materials with ordered pore structures and large
surface areas have shown great promise for use in industrial appli-
cations ranging from air to water purication. MCM-41, which is a
member of the M41S family of siliceous materials, is one popular
example of this type of structured mesoporous material. This
siliceous material, which was rst created by Mobil scientists in
the early 1990s [9], forms a hexagonal close-packed structure
composed of unidirectional channels arranged in a hexagonal man-
ner [10]. It has a high surface area and a repeating structure of
cylindrical pores, which is an ideal backbone for a well character-
ized adsorbent material.
Extensive research by Foley and others [1117] has provided
the foundation for the use of carbonized furfuryl alcohol as one
carbon phase in a composite material. Carbonized furfuryl alcohol
has been extensively studied as a carbonaceous adsorbent material
because it has a small pore size distribution centered at an average
pore size of 45 [11]. Pore formation by furfuryl alcohol depends
on the carbonization temperature [1113,15]. Burket et al. [11]
determined that both mesopores and micropores begin to appear
within the material at carbonization temperatures as low as
300 C. They proposed that mesopores are formed from the incom-
plete carbonization of polymer remnants and aromatic cores. As
the carbonization temperature was increased from 300 C to
600 C, the initial mesoporosity collapsed, leaving only micropores
ranging from 4 to 5 after carbonization at 500600 C. With such
1387-1811/$ - see front matter 2012 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.micromeso.2012.07.032

Corresponding author. Tel.: +1 615 343 1672; fax: +1 615 343 7951.
E-mail address: m.douglas.levan@vanderbilt.edu (M.D. LeVan).
Microporous and Mesoporous Materials 165 (2013) 4854
Contents lists available at SciVerse ScienceDirect
Microporous and Mesoporous Materials
j our nal homepage: www. el sevi er . com/ l ocat e/ mi cr omeso
a narrow and controlled pore size distribution, furfuryl alcohol-
based CMS materials are useful for separating two gaseous
compounds of different sizes [18]. Carbonized furfuryl alcohol
within the pores of a siliceous material has been studied for its
molecular sieving properties. De Clippel et al. [16,17] reported a
furfuryl alcohol-based carbon silica composite that showed
remarkable properties for separating linear and branched alkanes
due to the narrow pore size distribution provided by the impreg-
nated furfuryl alcohol phase. The small pore sizes and large surface
area of this CSC-furfuryl alcohol (CSC-FA) material make it a good
candidate for the removal of light gases for air purication [19].
Carbonized sucrose is another carbon phase that could be of
interest in a CSC material. Sucrose as a carbon phase is different
from carbonaceous furfuryl alcohol since the pores are larger,
and the temperature necessary for carbonization is much lower
than that of furfuryl alcohol. Exploring a CSC containing a carbon
phase that can be carbonized at a lower temperature than furfuryl
alcohol could be benecial for future work on these adsorbents,
since many functionalization approaches are best performed at
low temperatures. [20,21].
Sucrose has been shown to form a mesoporous carbon phase
with high surface areas (on the order of 1000 m
2
/g) and a pore size
of approximately 20 [22,23]. Extensive work on sucrose carbon-
ization has been performed at different temperatures and using
different synthesis procedures. Ting et al. [23] performed a
one-pot synthesis using sucrose catalyzed with sulfuric acid and
carbonized at 900 C. The resulting mesoporous carbon had a sur-
face area of 1200 m
2
/g and 44 pores. Peng et al. [24] synthesized
mesoporous carbonaceous materials at carbonization tempera-
tures ranging from 400 to 600 C. Zhuang and Yang [25] success-
fully produced carbonaceous spheres from aqueous sucrose
solutions carbonized under high pressure at 175 C. Zheng et al.
[26] produced CMK-3 type materials by taking advantage of low
temperature carbonization of sucrose in ethanol at 200 C in a high
pressure reactor. Banham et al. [22] templated sucrose into
hexagonal mesoporous silicas. Sucrose has also been templated
into silica gel and carbonized at 800 C [27]. Bimodal porous
carbons have been produced by impregnating silica spheres with
sucrose and carbonizing at high temperatures [28].
This research focuses on carbonaceous CSC materials as bipha-
sic adsorbents for adsorption of light acidic and basic gases.
Although not all acidic and basic gases will adsorb in the same
way, sulfur dioxide and ammonia are used as representative
acidic and basic gases, respectively, to understand the overall
trends associated with adsorption on these composites. It builds
on previous research from our group published in this journal
that introduced a CSC material based on MCM-41 and furfuryl
alcohol [1], optimized its synthesis [29], characterized its light
gas adsorption properties for carbon dioxide, nitrogen, methane,
and ethane [1,29], and optimized the silica phase of the CSC for
basic gas adsorption [21]. In this paper, the carbonization method
for CSC materials with two different carbon phase precursors, fur-
furyl alcohol (CSC-FA) and sucrose (CSC-S), is investigated, and
these composites are tested for their feasibility for air purication
applications. The CSC containing sucrose as the carbon phase was
synthesized using a novel low carbonization temperature synthe-
sis procedure. The furfuryl alcohol-based CSC was synthesized at
different carbonization temperatures, and the pore sizes and
structure are examined after carbonization at different tempera-
tures to increase surface area in the CSCs. The materials are
characterized using nitrogen adsorption isotherms, X-ray diffrac-
tion, thermogravimetric analysis, and tested for their ammonia
and sulfur dioxide adsorption capacities. To the best of our
knowledge, this is the rst study of a carbon silica composite
synthesized to target the adsorption of acidic and basic gases
for air purication.
2. Experimental methods
2.1. Materials
2.1.1. MCM-41
Tetramethylammonium hydroxide pentahydrate, TMAOH,
(97%), tetramethylammonium silicate solution, TMASi, (99.99%,
1520 wt.% in water), sulfuric acid (95.098.0%), and furfuryl alco-
hol (99%) were purchased from Sigma Aldrich. Hexadecyltrimethy-
lammonium chloride, CTAC, (25%) in water was purchased from
Pfaltz and Bauer. A solution of ammonium hydroxide (29 wt.%) in
water and CabOSil M5 were purchased from Fisher Scientic.
2.1.2. CSC-S
Sucrose was purchased from Fisher Scientic, and dry ethanol
(200 proof) was purchased from Pharmco-aaper.
2.1.3. CSC-FA
Furfuryl alcohol (99%) and toluene (99%) were purchased from
Sigma Aldrich.
2.2. MCM-41 synthesis
All CSC materials include MCM-41 as the silica phase. Hexago-
nally-ordered MCM-41 with a 37 pore was synthesized accord-
ing to the procedure described in a previous study [21].
2.3. CSC-S
2.3.1. CSC-S synthesis
The novel sucrose-impregnated CSC material was synthesized
using a low temperature carbonization technique. Equal parts eth-
anol and water were mixed in a Teon-lined Parr reactor at room
temperature. One gramof sucrose was added to the mixture, which
was then covered and stirred vigorously for thirty minutes. Next,
0.2 g of calcined MCM-41 was added to the mixture and stirred
for an additional 1 h. The Parr reactor was sealed and put into
the oven at 200 C for either 24 h (to produce CSC-1) or 48 h (to
produce CSC-2). After the reaction was complete, the Parr reactor
was removed from the oven and cooled to room temperature.
The CSC-S material was separated from the brown solution via
vacuum ltration and rinsed with distilled water. The CSC-S was
air dried overnight.
2.4. CSC-FA
2.4.1. CSC-FA synthesis
The furfuryl alcohol-impregnated carbon silica composite mate-
rial was synthesized using the procedure outlined by Glover et al.
[1]. After impregnation with the furfuryl alcohol phase, samples
were carbonized at temperatures of 300, 500, and 600 C, following
the method previously described [1].
2.5. Materials characterization
2.5.1. Textural characterization
Adsorption isotherms were measured using a Micromeritics
ASAP 2020 at 196 C with UHP nitrogen as the analysis gas.
Prior to measurement, approximately 0.1 g of each sample was
degassed with heating to 90 C and vacuum to 10 lbar. After
reaching 10 lbar, the samples were heated to 100 C under
vacuum for an additional 6 h. Density functional theory (DFT)
provided in the ASAP 2020 software was used to calculate pore
volumes and pore size distributions. Pore volumes reported
correspond to P=P
0
0:999.
A.M.B. Furtado et al. / Microporous and Mesoporous Materials 165 (2013) 4854 49
2.5.2. X-ray diffraction (XRD)
XRD spectra were used to conrm the long range structure of
the native and impregnated MCM-41 samples. The spectra were
measured using a Scintag X1h/h automated powder diffractometer
with Cu target, a Peltier-cooled solid-state detector, a zero back-
ground Si(510) support, and a copper X-ray tube as the radiation
source. Spectra were collected from 1.2 to 7 2h using a step size
of 0.02.
2.5.3. Thermogravimetric analysis (TGA)
Thermogravimetric analysis was performed on the CSC materi-
als to determine the amount of carbon impregnated into the MCM-
41 using a TA Instruments Q600 SDT, a simultaneous DSC-TGA.
Samples were heated in zero air from room temperature to
600 C using a ramp rate of 5 C per minute with an air ow rate
of 10 ml per min. To fully burn off the carbonaceous phase, the
samples were maintained at 600 C for 3 h.
2.5.4. Transmission electron microscopy (TEM)
High resolution TEM was performed on the CSC materials to
investigate the MCM-41 mesoporous network using a Philips
CM20 electron miscroscope operating at 80 kV. The samples were
prepared by dispersing approximately 0.1 g of CSC powder into
approximately 0.5 mL isopropanol via sonication for 1 min. The
dispersion was then placed onto Lacey carbon copper grids and
air dried.
2.5.5. Target gas capacity
Equilibrium capacities for room temperature light gas adsorp-
tion of both ammonia and sulfur dioxide were measured for all
samples using a breakthrough apparatus, a schematic of which
has been given by Furtado et al. [21,30]. Although measured under
dynamic conditions, full breakthrough capacities were measured at
low ow rates and are equilibrium capacities, as they end after the
feed concentration exits the bed [31,32]. The equilibrium capaci-
ties agree with equilibrium results obtained using a gravimetric
(Cahn) balance [30]. Prior to analysis, all samples were regenerated
under vacuum at 120 C for 2 h.
The ow rate of gas across the adsorbent bed was kept constant
at 1133 mg/m
3
for ammonia (1500 ppm in helium) and 1428 mg/
m
3
for sulfur dioxide (500 ppm in helium). The capacity of the
adsorbent material, n (mol ammonia or sulfur dioxide/kg adsor-
bent), was calculated via material balance using [21,32]
n
F
m
Z
1
0
c
0
c dt 1
where c
0
is the feed molar concentration, and c is the efuent con-
centration at time t. The volumetric ow rate of gas through the
adsorbent bed, F, was adjusted to yield a breakthrough time of
approximately 1 h. The mass of the sample, m, was approximately
10 mg and was contained in a small cylindrical adsorbent bed.
The capacities calculated using the breakthrough apparatus have a
standard deviation on the order of 3%, and different batches of
CSC are repeatable to within 5%.
3. Results
3.1. Carbon silica composites synthesis and characterization
3.1.1. CSC-S carbonized for different reaction times
Two novel CSC materials using sucrose as the carbon phase
were successfully produced by impregnating MCM-41 with su-
crose using different reaction times. Using thermogravimetric
analysis, the amount of carbon loaded into the MCM-41 scales with
the reaction time, as shown in Fig. 1. The initial mass loss corre-
sponds to the liberation of adsorbed water and the second gradual
mass loss represents the loss of carbon from the samples. Table 1
summarizes the amount of carbon loaded into each sample.
Nitrogen isotherms are shown in Fig. 2 for both CSC-S materials
and for the unimpregnated MCM-41. The parent MCM-41 exhibits
100
80
60
40
20
0
W
e
i
g
h
t

P
e
r
c
e
n
t
600 500 400 300 200 100 0
Temperature (K)
CSC-1
CSC-2
Fig. 1. Thermogravimetric analysis of CSC-S samples.
Table 1
Physical properties of the CSC-S samples.
Sample Reaction
time (h)
wt.% C BET SA
(m
2
/g)
V
pore
(cm
3
/g)
Predominant
pores ()
MCM-41 836 1.2 37
CSC-1 24 26 436 0.56 1230
CSC-2 48 46 263 0.45 1230
700
600
500
400
300
200
100
0
A
m
o
u
n
t

A
d
s
o
r
b
e
d

(
c
m
3
/
g

a
t

S
T
P
)
1.0 0.8 0.6 0.4 0.2 0.0
Relative Pressure (P/P
0
)
MCM-41
CSC-1
CSC-2
Fig. 2. Nitrogen adsorption isotherms for CSC-S samples.
50 A.M.B. Furtado et al. / Microporous and Mesoporous Materials 165 (2013) 4854
Type IV behavior according to the IUPAC classication [34]. The
CSCs are microporous and mesoporous, and exhibit Type I behavior
due to the sharp increase in the amount adsorbed at low relative
pressures, which levels off as the relative pressure increases. BET
surface areas and pore volumes calculated from the isotherms
are summarized in Table 1. Impregnating MCM-41 with the
sucrose phase decreases the surface area and pore volume of the
materials compared to the parent MCM-41, as shown by the col-
umn in Table 1 summarizing the predominant pores in the samples
and by the pore size distributions shown in Fig. 3. This is due to
lling the mesopores with a carbon phase, which results in an
increase in microporosity and a decrease in mesoporosity. After
lling the MCM-41 pores with different amounts of sucrose carbon,
density functional theory predicts lower predominant pore sizes
compared to the initial 37 pore of the parent MCM-41. Conse-
quently, it is evident that sucrose carbon lls the MCM-41 mesop-
ores and produces micropores and small mesopores in the material
after carbonization.
High resolution TEM images were obtained to compare MCM-
41 to the CSC-S samples. Fig. 4 compares TEM images of the parent
MCM-41 with CSC-1. In image a, which shows the parent MCM-41,
the pores are clearly visible. Image analysis using the Philips CM20
software calculated the pore diameters to be 38.5 , which com-
pares well with the expected value of 37 . Image b shows the
aligned sucrose carbon-impregnated MCM-41 pores of 38.5 in
diameter. Also apparent during analysis of this CSC-1 sample were
carbon spheres ranging from 500 nm to 1 micron in diameter.
These spheres, shown in image c alongside the aligned CSC-S
cylindrical pores, are similar to those found by Zhuang and Yang
[25] in their low temperature carbon synthesis, and are structures
forming around carbon lled MCM-41 pores.
3.1.2. CSC-FA carbonized at different temperatures
The CSC-FA materials were characterized by X-ray diffraction to
verify that the impregnation of the silica phase with carbon did not
disrupt the MCM-41 structure. Fig. 5 shows the XRD spectra of the
base CSC-FA materials carbonized at different temperatures and
the MCM-41 base material. Despite the decrease in intensity of
the base CSC materials compared to MCM-41, which is due to
interference from the carbon phase, the MCM-41 peaks are identi-
able in the CSC-FA spectra. Thus, carbon impregnation does not
alter the MCM-41 hexagonal order. Long range XRD scans were
also run to identify evidence of graphitization, which can be
expected to occur during carbonization at higher temperatures.
No graphitic peaks were detected in these scans.
Characterization of the CSC-FA materials was performed by
measuring nitrogen isotherms at 77 K. These are shown in Fig. 6
and are Type I isotherms. The nitrogen isotherm for the CSC300
sample is not shown in Fig. 6 because it adsorbs much less nitrogen
than the other samples and is on a much lower scale. All CSC-FA
materials are nanoporous and exhibit Type I behavior due to the
sharp increase in the amount adsorbed at low relative pressures,
which levels off as the relative pressure approaches unity. The
8
6
4
2
0
D
i
f
f
e
r
e
n
t
i
a
l

P
o
r
e

V
o
l
u
m
e

(
c
m
3
/
g
)
35 30 25 20 15 10
Pore Width ()
MCM-41
1.0
0.8
0.6
0.4
0.2
0.0
D
i
f
f
e
r
e
n
t
i
a
l

P
o
r
e

V
o
l
u
m
e

(
c
m
3
/
g
)
35 30 25 20 15 10
Pore Width ()
CSC-1
CSC-2
CSC300
CSC500
CSC600
Fig. 3. Pore size distributions calculated via DFT for MCM-41, CSC-S, and CSC-FA
samples. The curve for CSC300 lies just above the x-axis.
Fig. 4. High resolution transmission electron microscopy images of MCM-41 and CSC-1.
A.M.B. Furtado et al. / Microporous and Mesoporous Materials 165 (2013) 4854 51
hysteresis during desorption is characteristic of a Type IV isotherm
[19].
Surface area and pore information were calculated from the
nitrogen adsorption isotherms of the CSC-FA materials. BET surface
areas were calculated according to the procedure for microporous
materials outlined by Rouquerol et al. [33] and are summarized in
Table 2 with pore size distributions given in Fig. 3. When all sam-
ples are compared, those carbonized at mid-range temperatures
(500600 C) have much higher surface areas and larger pore vol-
umes than the sample carbonized at 300 C, which has minimal
porosity. Also, as shown in Fig. 3, when compared with the su-
crose-based materials, the FA-based materials have pore size dis-
tributions shifted more towards the micropore region.
TGA analysis was performed on the CSC-FA samples produced
at different carbonization temperatures. As summarized in Table 2,
data show that the CSC-FA carbonized at 300 C has the highest
mass loss, 51 wt.%, due to incomplete polyfurfuryl alcohol carbon-
ization at the lower temperature [1113,15]. The average carbon
content for the remaining samples is approximately 40%.
High resolution TEM images were also obtained on these sam-
ples. Fig. 7 compares TEM images of MCM-41 and CSC500, a repre-
sentative CSC-FA sample. Image a shows the parent MCM-41
material. Image b shows the carbon-lled MCM-41 arranged in
an ordered manner. In image c, the hexagonal pores are visible. It
is difcult to distinguish between the carbon lled MCM-41 pores
and the base MCM-41. It is evident from these images that the fur-
furyl alcohol polymerizes well within the MCM-41 mesopores
rather than forming a pure carbon phase outside of the MCM-41.
3.2. CSC target gas adsorption
3.2.1. CSC-S
As shown in Table 3, the sucrose-based carbon silica composites
were tested for ammonia and sulfur dioxide adsorption capacity
and compared to MCM-41 and BPL activated carbon. This table in-
cludes two columns presenting ammonia and sulfur dioxide
adsorption capacities calculated per kg sample, and two columns
presenting these capacities calculated on a per kg silica and carbon
basis, respectively. Calculating the capacities per kg of each phase
was done to emphasize that the carbon phase in the composite tar-
gets sulfur dioxide adsorption and the silica phase targets ammo-
nia adsorption. It should be emphasized that the formation of the
composite gives a material with a different pore size distribution
and surface chemistry than either the parent MCM-41 or a su-
crose-based carbon created outside of the MCM-41 mesopores.
The pore size distribution of the carbon phase can impact the
adsorption of ammonia targeted for the silica phase and the tem-
plating silica support can impact the carbon phase targeted for sul-
fur dioxide adsorption. Thus, there are synergistic interactions in
the formation of the two phases of the composite material for
adsorption of the two target gases.
It is obvious from the table that the presence of the sucrose car-
bon phase increases both the ammonia and sulfur dioxide adsorp-
tion capacities compared to the parent MCM-41. The MCM-41
provides hydroxyl groups that enhance ammonia adsorption
through hydrogen bonding [21,3537], and these sites provide
much of the ammonia adsorption capacity of the CSC material.
The parent MCM-41 has minimal sulfur dioxide adsorption capac-
ity. However, after impregnation, the carbon phase introduces
micropores into the composite that promote adsorption of sulfur
dioxide and also ammonia. The microporous carbon phase en-
hances the sulfur dioxide capacity of the composite by 400% and
it enhances the ammonia adsorption capacity by 9%. The drastic in-
crease in sulfur dioxide capacity after impregnation results from
the basic nature of the sucrose carbon phase [38,39], thereby
enhancing the acidic gas adsorption of the composite. The much
smaller increase in ammonia adsorption capacity is due to the in-
creased presence of micropores throughout the sample. Compared
to BPL activated carbon, both CSC samples show much higher
I
n
t
e
n
s
i
t
y

(
c
o
u
n
t
s

p
e
r

s
e
c
o
n
d
)
7 6 5 4 3 2
Degrees 2
MCM-41
CSC300
CSC500
CSC600
Fig. 5. X-ray diffraction spectra for MCM-41 and CSC-FAs carbonized at different
temperatures.
700
600
500
400
300
200
100
0
A
m
o
u
n
t

A
d
s
o
r
b
e
d

(
c
m
3
/
g

a
t

S
T
P
)
1.0 0.8 0.6 0.4 0.2 0.0
Relative Pressure (P/P
0
)
MCM-41
CSC500
CSC600
Fig. 6. Nitrogen adsorption isotherms for CSC-FAs carbonized at different
temperatures.
Table 2
Physical properties of the CSC-FA samples.
Sample Carbonization
temperature (K)
wt.% C BET SA
(m
2
/g)
V
pore
(cm
3
/g)
Predominant
pores ()
MCM-41 836 1.2 35
CSC300 300 51 24 0.07 Minimal
CSC500 500 40 466 0.39 820
CSC600 600 44 505 0.42 825
52 A.M.B. Furtado et al. / Microporous and Mesoporous Materials 165 (2013) 4854
ammonia capacities. For sulfur dioxide, the carbon phase provides
base capacity for the composite, whereas the silica phase does not.
On a mol/kg carbon basis, the sulfur dioxide adsorption capacity
for both CSC-S materials is higher than that of BPL AC.
3.2.2. CSC-FA
Target gas adsorption capacities are shown in Table 4 for the
CSC-FA samples, the parent MCM-41, and BPL activated carbon.
The presence of the carbon phase enhances the sulfur dioxide
adsorption compared to the parent MCM-41, with CSC600 having
the highest SO
2
adsorption capacity. Similar to the sucrose carbon
phase, the carbon phase produced by the furfuryl alcohol is basic in
nature [38,39], and it enhances the sulfur dioxide adsorption
capacity over that of the parent MCM-41. When compared on a
mol/kg silica basis, the carbon phase produced at the 300 C car-
bonization temperature causes a decrease in the ammonia adsorp-
tion capacity compared to MCM-41; however, the carbon phases in
CSC500 and CSC600 result in an increase in the ammonia adsorp-
tion capacity compared to the base MCM-41. For the CSC-FA sam-
ples, it is obvious that higher adsorption capacities correspond to
higher carbonization temperatures. The nitrogen isotherms of
Fig. 6 show much higher surface areas and well developed pore
structures for the samples carbonized at 500 and 600 C compared
to the sample carbonized at the lower temperature. Analyses of the
samples show incomplete furfuryl alcohol carbonization at the low
carbonization temperature of 300 C, giving a decrease in ammonia
and sulfur dioxide adsorption capacity compared to the samples
carbonized at 500 and 600 C. All CSC samples show much higher
ammonia adsorption capacities than BPL activated carbon. On a
mol/kg carbon basis, all CSC materials also have higher sulfur diox-
ide adsorption capacities than the BPL activated carbon. When
compared on a mol/kg sample basis, CSC500 and CSC600 show
higher sulfur dioxide adsorption capacities than the commercial
carbon.
4. Conclusions
A series of carbon silica composites with MCM-41 as the silica
phase and carbonized sucrose or furfuryl alcohol as the carbon
phase have been synthesized using different carbonization temper-
atures and reaction times. These materials have been characterized
via adsorption isotherms, XRD, and TGA. They were also tested for
their ammonia and sulfur dioxide capacities, a basic and an acidic
gas, respectively.
Impregnation of MCM-41 with carbonized sucrose results in a
novel CSC material with potential as an air purication adsorbent.
CSC-S shows an increase in the sulfur dioxide and ammonia
adsorption capacities compared to the unimpregnated MCM-41.
The carbonized sucrose phase of the CSC-S composite enhances
the sulfur dioxide adsorption capacity of the adsorbent compared
to the unimpregnated silica phase. The presence of the sucrose
phase also results in an increase in ammonia adsorption capacity.
The increases in ammonia and sulfur dioxide adsorption capacities
occur for samples impregnated using both 24 and 48 h reaction
times.
Similar to the CSC-S material, impregnation of MCM-41 with
furfuryl alcohol to form the CSC-FA results in a composite with
high toxic gas capacities. The CSC-FA maintains the ammonia
adsorption capacity and enhances the sulfur dioxide adsorption
capacity compared to MCM-41. The CSC-FA carbonized at 600 C
has the highest surface area, followed by the samples carbonized
at 500 and 300 C. Carbonization of the furfuryl alcohol polymer
is a rate process, and the sample heated to 300 C is not fully car-
bonized, whereas the sample carbonized at 600 C has the most
well-developed pore structure. The development of the pore struc-
ture correlates with the ammonia and sulfur dioxide capacities; in
general, the samples carbonized at higher temperatures have
higher target gas capacities than the sample carbonized at 300 C.
Carbon silica composites produced by impregnating MCM-41
with sucrose and furfuryl alcohol carbons show promise as adsor-
bent materials for air purication. These materials provide a car-
bon phase for physical adsorption and a silica phase for hydrogen
Fig. 7. High resolution transmission electron microscopy images of MCM-41 and CSC500.
Table 3
Gas adsorption on the CSC-S samples.
Sample SO
2
adsorption capacity NH
3
adsorption capacity
mol/kg sample mol/kg carbon mol/kg sample mol/kg SiO
2
BPL AC 0.20 0.20 0.10
MCM-41 0.02 2.01 2.01
CSC-1 0.10 0.38 2.18 2.95
CSC-2 0.15 0.33 2.10 3.93
Table 4
Gas adsorption on the CSC-FA samples.
Sample SO
2
adsorption capacity NH
3
adsorption capacity
mol/kg sample mol/kg carbon mol/kg sample mol/kg SiO
2
BPL AC 0.20 0.20 0.10
MCM-41 0.02 2.01 2.01
CSC300 0.20 0.55 0.76 1.19
CSC500 0.25 0.71 1.38 2.14
CSC600 0.34 0.95 1.48 2.30
A.M.B. Furtado et al. / Microporous and Mesoporous Materials 165 (2013) 4854 53
bonding with electronegative atoms. These biphasic materials have
large capacities for two very different types of gases, ammonia and
sulfur dioxide. Their adsorption capacities for these gases are
higher than those of BPL activated carbon.
Acknowledgements
We are grateful to the US Army Edgewood Chemical and Biolog-
ical Center and the Defense Threat Reduction Agency for the sup-
port of this research under contract number W911SR-08-C-0028.
We are also grateful to Dr. James McBride for his help with the
transmission electron microscope.
References
[1] T.G. Glover, K.I. Dunne, R.J. Davis, M.D. LeVan, Micropor. Mesopor. Mater. 111
(2008) 111.
[2] F. Rodriguez-Reinoso, M. Molina-Sabio, M.A. Muiiecas, J. Phys. Chem. 96 (1992)
27072713.
[3] W.A. Noyes, Military Problems with Aerosols and Nonpersistent Gases,
Summary Technical Report of Division 10, National Defense Research
Committee, Washington, DC, 1946.
[4] Department of Health and Human Services, National Institutes of
Occupational, Safety and Health, 2003.
[5] H. Fortier, P. Westreich, S. Selig, C. Zelenietz, J.R. Dahn, J. Colloid Interface Sci.
320 (2008) 423435.
[6] T.J. Bandosz, J. Jagiello, K. Putyera, J.A. Schwarz, Langmuir 11 (1995) 3964
3969.
[7] N. Sonobe, T. Kyotani, A. Tomita, Carbon 28 (1990) 483488.
[8] T.J. Bandosz, J. Jagiello, K. Amankwah, J.A. Schwarz, Clay Miner. 27 (1992) 435
444.
[9] C.T. Kresge, M.E. Leonowicz, W.J. Roth, J.C. Vartuli, J.S. Beck, Nature 359 (1992)
710712.
[10] B. Marler, U. Oberhagemann, S. Vortmann, H. Gies, Micropor. Mater. 6 (1996)
375383.
[11] C.L. Burket, R. Rajagopalan, A.P. Marencic, K. Dronvajjala, H.C. Foley, Carbon 44
(2006) 29572963.
[12] S. Bertarione, F. Bonino, F. Cesano, A. Damin, D. Scarano, A. Zecchina, J. Phys.
Chem. B 112 (2008) 25802589.
[13] H.C. Foley, Micropor. Mater. 4 (1995) 407433.
[14] K. Dettmer, W. Engewald, Anal. Bioanal. Chem. 373 (2002) 490500.
[15] D. Lafyatis, J. Tung, H.C. Foley, Ind. Eng. Chem. Res. 30 (1991) 865873.
[16] F. de Clippel, A. Harkiolakis, X. Ke, d T. Vosch, G. Van Tendeloo, G.V. Baron, P.A.
Jacobs, J.F.M. Denayer, B.F. Sels, Chem. Commun. 46 (2010) 928930.
[17] F. de Clippel, A. Harkiolakis, T. Vosch, X. Ke, L. Giebeler, S. Oswald, K.
Houthoofd, J. Jammaer, G. Van Tendeloo, J.A. Martens, P.A. Jacobs, G.V. Baron,
B.F. Sels, J.F.M. Denayer, Micropor. Mesopor. Mater. 144 (2011) 120133.
[18] A. Hong, R.K. Mariwala, M.S. Kane, H.C. Foley, Ind. Eng. Chem. Res. 34 (1995)
992996.
[19] F. Rouquerol, J. Rouquerol, K. Sing, Adsorption by Powders and Porous Solids,
Academic Press, San Diego, 1999.
[20] N.R.E.N. Impens, P. van der Voort, E.F. Vansant, Micropor. Mesopor. Mater. 28
(1999) 217232.
[21] A.M.B. Furtado, Y. Wang, T.G. Glover, M.D. LeVan, Micropor. Mesopor. Mater.
142 (2011) 730739.
[22] D. Banham, F. Feng, J. Burt, E. Alsrayheen, V. Birss, Carbon 48 (2010) 1056
1063.
[23] C. Ting, H. Wub, S. Vetrivel, D. Saikia, Y. Pan, G.T.K. Fey, H. Kao, Micropor.
Mesopor. Mater. 128 (2010) 111.
[24] L. Peng, A. Philippaerts, X. Ke, J. Van Noyen, F. De Clippel, G. Van Tendeloo, P.A.
Jacobs, B.F. Sels, Cat. Today 150 (2010) 140146.
[25] Z. Zhuang, Z. Yang, J. App. Pol. Sci. 114 (2009) 38633869.
[26] M. Zheng, Y. Liu, K. Jiang, Y. Xiao, D. Yuan, Carbon 48 (2010) 12241233.
[27] A. Puziy, O.I. Poddubnaya, C.A. Reinish, M.M. Tsyba, L.I. Mikhalovska, S.V.
Mikhalovsky, Carbon 49 (2011) 599604.
[28] S. Zhang, L. Chen, S. Zhou, D. Zhao, L. Wu, Chem. Mater. 22 (2010) 34333440.
[29] T.G. Glover, M.D. LeVan, Micropor. Mesopor. Mater. 118 (2009) 2127.
[30] A.M.B. Furtado, J. Liu, Y. Wang, M.D. LeVan, J. Mater. Chem. 21 (2011) 6698
6706.
[31] A. Kizzie, A.G. Wong-Foy, A.J. Matzger, Langmuir 27 (2011) 63686373.
[32] R.H. Perry, D.W. Green (Eds.), Chapter 16: Adsorption and Ion Exchange,
Perrys Chemical Engineers Handbook, 8th ed., McGraw Hill, New York, 2008.
[33] J. Rouquerol, P. Llewellyn, F. Rouquerol, Stud. Surf. Sci. Catal. 160 (2007) 4956.
[34] S.J. Gregg, K.S.W. Sing, Adsorption, Surface Area and Porosity, Academic Press,
London, 1982.
[35] B.A. Morrow, I.A. Cody, L.S.M. Lee, J. Phys. Chem. 79 (1975) 24052408.
[36] B.A. Morrow, I.A. Cody, J. Phys. Chem. 80 (1976) 19982004.
[37] S. Kittaka, M. Morimura, S. Ishimaru, A. Morino, K. Ueda, Langmuir 25 (2009)
17181724.
[38] D. Stosic, S. Bennici, J. Couturier, J. Dubois, A. Auroux, Cat. Commun. 17 (2012)
2328.
[39] M. Seredych, T.J. Bandosz, J. Phys. Chem. C 111 (2007) 1559615604.
54 A.M.B. Furtado et al. / Microporous and Mesoporous Materials 165 (2013) 4854