Co-MCM-41 for Production of SWNTs
J. Phys. Chem. B, Vol. 107, No. 40, 2003 11049
Experimental Section
relationship between the lattice parameter and the spacing. (ao
) 2d100/x3) and N2 physisorption results for pore diameter
determination. The wall thickness was estimated to be 1 nm,
independent of the pore diameter.
DR UV-Vis. The UV-vis spectra were recorded by diffuse
reflectance on a Hewlett-Packard 8452A diode array spectrom-
eter equipped with a Harrick praying mantis. All spectra were
recorded at room temperature under ambient atmosphere.
Samples consisting of 100 mg of powder were pressed in the
sample holder by hand to make a thick wafer. The reference
samples were diluted to 1 wt % with siliceous MCM-41.The
final spectra of cobalt were obtained by subtracting the spectra
of the pure siliceous MCM-41.
X-ray Absorption. X-ray absorption measurement was per-
formed at the Co K edge (7709 eV) using Si (111) as the
monochromator crystal at station X23A2 in NSLS, 2.5 GeV
storage ring, Brookhaven National Laboratory. Samples were
pressed into self-supporting wafers and placed in a stainless
steel cell equipped with water-cooled Kapton windows, a gas
inlet and outlet, and a heating unit allowing in situ gas treatment.
Details on the experimental procedure are given elsewhere.10
Raman Spectroscopy. Raman spectra were recorded on a
LabRam instrument from Jobin Yvon Horiba equipped with an
Olympus confocal microscope using an excitation wavelength
of 532 nm.
Materials. Silica synthesis sources used were HiSil-915 from
Pittsburgh Plate Glass (PPG) and tetramethylammonium silicate
(10 wt % silica, SACHEM Inc.). The Co source was CoSO4‚
xH2O (Aldrich Chemical Co.). The quaternary ammonium
surfactants CnH2n+1(CH3)3NBr were purchased from Aldrich
Chemical Co. with n ) 12, 14, 16, 18 and from American Tokyo
Kasei with n ) 10. The surfactant solutions were prepared by
ion-exchanging the 29 wt % (C10 and C12), 20 wt % (C14 and
C16), and 10 wt % (C18) CnH2n+1(CH3)3NBr aqueous solutions
with equal molar exchange capacity of Amberjet-400 (OH) ion-
exchange resin (Sigma Chemical Co.) by overnight batch
mixing. The antifoaming agent was Antifoam A from Sigma
Chemical Co., a silane polymer alkyl terminated by methoxy
groups. Acetic acid (Fisher Scientific) was used for pH
adjustment of the synthesis solution. In the text hereafter specific
samples are designated by the alkyl chain length of the surfactant
used (e.g., C10 Co-MCM-41).
Synthesis. HiSil-915, tetramethylammonium silicate and the
Co aqueous solution were mixed for 30 min. The water to total
silica ratio was varied from 74.4 to 86.0 mole ratio based on
the surfactant chain length. The surfactant solution was added
to the prepared silica and Co mixture and a small amount of
antifoaming agent (0.2 wt % of surfactant) was incorporated to
remove excess foam produced by the surfactant as a result of
stirring the synthesis solution. Acetic acid was added until pH
) 11 was reached. After additional mixing for about 10 min,
this synthesis solution was poured into a polypropylene bottle
and placed in the autoclave at 100 or 150 °C for 6 days. After
cooling to room temperature, the resulting solid was recovered
by filtration, washed with deionized water and dried under
ambient conditions. The predried solid was heated from room
temperature to 540 °C for 20 h under He (30 mL/min) and
soaked for 1 h at 540 °C in flowing He followed by 6 h of
calcination at 540 °C under flowing air to remove the residual
organics. The molar ratio of each component in the synthesis
solution was controlled at SiO2: surfactant:Co:H2O ) 1:0.27:
0.01:X (X ) 74.4-86.0). Because the preparation process may
cause some loss of Co and silica in the byproducts, the final
Co content of each sample was determined by ICP at Galbraith
Laboratories, Inc. A pure siliceous MCM-41 was also prepared
following the procedure described above for Co-MCM-41,
without incorporating the cobalt into the synthesis solution.
Characterizations. N2 Physisorption. Nitrogen adsorption-
desorption isotherms were measured at -196 °C with a static
volumetric instrument Autosorb-1C (Quanta Chrome). Prior to
the measurement, the samples were outgassed at 200 °C to a
residual pressure below 1 × 10-4 Torr. A Baratron pressure
transducer (0.001-10 Torr) was used for low-pressure measure-
ments. The pore size distributions were calculated from the
desorption isotherms using the BJH method.8 Ravikovitch et
al.9 reported that pore diameter values determined using the BJH
method are underestimated, and the nonlocal density functional
theory (NLDFT) is more reliable for the mesopore size
prediction. However, the pore size distribution determined in
our study by the BJH method provides reliable results that can
be used for the relative comparison of the synthesized samples.
Other complementary techniques such as HR-TEM were also
used to confirm the structure and the pore size.
Results and Discussion
Fresh Co-MCM-41 Samples (before Reaction). Figure 1
shows the nitrogen physisorption results of C10 to C18 Co-
MCM-41 samples. In this study, nitrogen physisorption was used
as a standard method to compare the structure of each sample
because it shows the volume averaged value, unlike XRD or
TEM, which only probe a limited part of the samples. The full
width at half-maximum (fwhm) of the pore size distribution
(PSD) curve and the slope of the capillary condensation in the
isotherms were determined for each sample and used as structure
indexes for comparisons. Figure 1a shows the isotherms of C10
to C18 Co-MCM-41 samples. All samples show capillary
condensation, suggesting the Co-MCM-41 samples were suc-
cessfully synthesized regardless of the surfactant chain length.
As is well-known, the longer surfactant chain length results in
the higher relative pressure of the capillary condensation. When
the pore size distribution for each sample was compared, the
pore size increased systematically and the corresponding pore
volume varied linearly with the surfactant chain length. This
implies that the pore size and the pore volume can be precisely
controlled using the synthesis method described in this study.
The mesopore volume, defined in this study as the volume of
pores with diameters below 10 nm, shows a linear dependence
on the surfactant chain length (see Figure 2). This control shows
potential for controlling the size and the amount of carbon
nanotubes produced in the pore system of these catalytic
template materials. The slope of capillary condensation and the
fwhm used as structure indexes are listed in Table 1. Longer
surfactant chain lengths result in narrower fwhms and steeper
slopes for the capillary condensation. These results are highly
reproducible with most of the transition metal incorporated
MCM-41 samples prepared following the method described in
the Experimental Section. Modeling work aimed at predicting
the sample structure and metal loading is under development
for V-MCM-41,11 and it will be extended to Co-MCM-41
soon.
XRD. X-ray diffraction measurements were carried out using
a Shimadzu X-ray diffractometer (Cu KR, λ ) 0.154 nm, 40
kV, 30 mA) to check if the prepared Co-MCM-41 has the
characteristic hexagonal pore structure after calcination and
reaction. The pore wall thickness was calculated from the XRD
The sample synthesized using the C18 surfactant shows a
broader fwhm than those of Co-MCM-41 samples synthesized