- Email: [email protected]
Synthesis of mesoporous carbons using silica templates impregnated with mineral acids
Accepted Manuscript Synthesis of Mesoporous Carbons using Silica Templates Impregnated with Mineral Acids Yongbeom Seo, Kyoungsoo Kim, Younjae Jung, Ryong Ryoo PII:
S1387-1811(15)00033-5
DOI:
10.1016/j.micromeso.2015.01.021
Reference:
MICMAT 6946
To appear in:
Microporous and Mesoporous Materials
Received Date: 2 October 2014 Revised Date:
7 January 2015
Accepted Date: 15 January 2015
Please cite this article as: Y. Seo, K. Kim, Y. Jung, R. Ryoo, Synthesis of Mesoporous Carbons using Silica Templates Impregnated with Mineral Acids, Microporous and Mesoporous Materials (2015), doi: 10.1016/j.micromeso.2015.01.021. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Synthesis of Mesoporous Carbons using Silica Templates
RI PT
Impregnated with Mineral Acids Yongbeom Seo,† Kyoungsoo Kim,‡ Younjae Jung,† and Ryong Ryoo‡,†,*
Department of Chemistry, Korea Advanced Institute of Science and Technology (KAIST), Daejeon 305-701, Korea
‡
Center for Nanomaterials and Chemical Reactions, Institute for Basic Science (IBS), Daejeon 305-701, Korea
*
Corresponding author: Ryong Ryoo (E-mail: [email protected]; Tel.: +82 42 350 2830)
M AN U
SC
†
Abstract
The pore walls of SBA-15, SBA-16 and KIT-6 mesoporous silica were acidified via the impregnation of phosphoric acid or sulfuric acid prior to their use as a template for mesoporous carbon synthesis with furfuryl alcohol. The effect of acid functionalization on carbon synthesis was investigated using X-ray powder diffraction, N2 adsorption, transmission electron microscopy, and solid-state 13C and 31P NMR spectroscopy. The results indicated that a
TE D
significant portion of the impregnated mineral acid formed an ester with the surface silanol group. Therefore, the impregnated acid could lead to the selective formation of carbon frameworks inside the template pores, preventing external carbon deposition even when furfuryl alcohol was used excessively. The mineral acid impregnation yielded
EP
highly faithful carbon replicas of the mesoporous silica.
AC C
Keywords: template synthesis; mesoporous material; ordered mesoporous carbon; mineral acid; solidstate NMR
1
ACCEPTED MANUSCRIPT
1. Introduction Mesopores are defined as pores in the range of 2 to 50 nm in diameter. Carbon materials with regular arrays of these mesopores are called ordered mesoporous carbons (OMCs) [1-8]. OMCs with various structures are currently
RI PT
available via a synthetic route using mesoporous silica templates [1-6] or carbonization of ordered arrays of
polymeric resin nanobeads [7, 8]. In the current work, we focus on the template synthesis. The template synthesis of
OMCs was first reported by Ryoo et al. in 1999 [1]. These types of OMCs are referred to as ‘CMK’. The CMK-type of carbon materials have attracted much attention as absorbents, catalyst supports, and electrode materials [3, 9-12].
SC
There have been numerous reports on the synthesis, characterization and application of CMK carbons with various
structures and pore diameters [2-6, 9-14]. Surface modification and functionalization of CMK carbons have also been extensively studied to improve their performance [15-18].
M AN U
The synthesis procedure for CMK carbons involves three steps [1-5]. In the first step, an organic compound is impregnated into the mesopores of a silica template. The impregnated organic source is polymerized and subsequently converted to carbon by pyrolysis. The key in this step is to control the carbonization process to occur inside the template mesopores. The confinement of the carbonization process within the mesopores can be realized by converting the organic source to cross-linked solid polymers prior to the final conversion to carbon. The polymerization can be achieved by the incorporation of an acid into the silica template or the carbon source. The
TE D
silica template can be removed after carbonization with hydrofluoric acid or a sodium hydroxide solution. Ultimately, the carbon product should become a faithful inverse replica of the porous silica template. Various organic compounds, such as sucrose [1, 2, 5], furfuryl alcohol (FA) [3], acenaphthene [19], phenol resin [20], and acrylonitrile [15], can be used as the carbon source. FA is one of the most convenient sources of carbon as
EP
the liquid-phase compound that can be easily impregnated into silica mesopores. Nevertheless, the carbonization process using FA involves several time consuming steps. For example, incorporation of a solid acid (typically Al) onto the template pore walls is necessary. The procedure involves wet impregnation of an Al compound and
AC C
evaporation of the solvent followed by calcination. Then, FA is impregnated using an incipient wetness process. The
impregnated FA is polymerized by heating at 373 K due to the catalytic function of the acid sites. Upon further heating to 623 K, the carbon source is converted to a black tar with a significant contraction in the volume. An
additional amount of FA can be added to compensate for the volume loss (i.e., second impregnation) at this point.
Then, the procedures for FA polymerization at 373 K and subsequent tar formation are repeated. After all of these treatments, the silica template containing the carbon source is heated to the final carbonization temperature (~ 1173 K). The resultant carbon from the two-step FA impregnation exhibits a rod-type framework that fills the template pores. If the second FA impregnation is skipped, the carbon often becomes a hollow pipe-type framework [3].
2
ACCEPTED MANUSCRIPT
It is important not to exceed the amount of FA beyond the total pore volume in a template. An excessive amount of FA can cause the formation of non-templated (i.e., external) carbon. The acid strength of the template pore walls is also an important factor. Often the polymerization rate of carbon sources including FA, sucrose, glucose, and other
RI PT
organics are very sensitive to acid catalysts [21 - 27]. For the aluminosilicate template, the acid strength is often so high that FA polymerization proceeds very rapidly before a sufficient time can allow for homogeneous infiltration of
FA into the template pores. The rapid polymerization can be a serious problem particularly in a large batch synthesis. The exothermic polymerization process can generate substantial heat that accelerates the polymerization itself [21,
SC
22]. The FA polymerization outside the template can result in the formation of external carbon. There were
alternative routes involving the use of silica templates without Al incorporation [23 - 27]. In such cases, a small amount of p-toluene sulfonic acid or oxalic acid were added to the FA as a polymerization catalyst prior to the FA
M AN U
being impregnated. This route is prone to suffering from the generation of polymerization heat as well as the formation of external carbon.
The current work was undertaken to improve the synthesis procedure for OMC materials using FA in mesoporous silica templates. In particular, this study was aimed at preventing the generation of external carbon by selecting a moderately weak acid that would slowly catalyze the FA polymerization. As a first candidate, phosphoric acid was impregnated into ordered mesoporous silica prior to FA infiltration. The resulting carbonization process was
TE D
investigated using X-ray powder diffraction, N2 adsorption, and transmission electron microscopy. The chemical state of the phosphoric acid in the silica template was analyzed by solid-state 31P NMR spectroscopy. Based on the chemical information from the 31P NMR data, we were able to extend the phosphoric acid-based method to the use of sulfuric acid. In this paper, we introduce a mineral acid impregnation method, and its application to the synthesis of
EP
OMCs with various mesoporous silica templates, such as SBA-15, KIT-6 and SBA-16.
2. Experimental
AC C
2.1. Synthesis
2.1.1. Synthesis of silica templates SBA-15 silica was synthesized according to the procedure described in the literature [29] using tetraethoxysilane
(TEOS, TCI) as a silica source and Pluronic P123 [EO20PO70EO20 (EO = ethylene oxide, PO = propylene oxide, Mn ~5800, Aldrich] as the mesopore-directing surfactant. The gel composition was 1 TEOS: 0.017 Pluronic P123: 5.3 HCl: 194 H2O. All of the starting materials were mixed at 308 K for 12 h using a magnetic stirrer. Then, the resulting
gel mixture was heated for 12 h at 403 K under static conditions in a Teflon-lined autoclave. The precipitated SBA-
3
ACCEPTED MANUSCRIPT
15 product was filtered, washed with deionized water and dried at 373 K for 12 h. The product was calcined in air at 823 K for 4 h. SBA-16 and KIT-6 were synthesized according to previously reported procedures [6, 30].
RI PT
2.1.2. OMC synthesis using aluminated SBA-15 In a typical procedure for post-synthetic alumination of SBA-15 with Si/Al of 20, 0.5 g of the calcined SBA-15
silica was added to 20 mL of ethanol (99.9%, Merck) containing 0.055 g of AlCl3 (anhydrous, Aldrich). The mixture was magnetically stirred at room temperature for 6 h. Then, ethanol was removed using a rotary evaporator. The
SC
SBA-15 containing Al (i.e., Al-SBA-15) was further dried in an oven at 373 K for 3 h followed by calcination in an air stream at 823 K for 4 h. The inductively coupled plasma-atomic emission spectroscopy (ICP-AES) analysis indicated that the Si/Al ratio of Al-SBA-15 was 22.
M AN U
An ordered mesoporous carbon (OMC) sample was prepared using Al-SBA-15 as a template and FA (98%, Aldrich) as a carbon precursor. In a typical synthesis, 0.1 g of Al-SBA-15 was impregnated with 0.14 mL (amounting to 130% of the total pore volume of Al-SBA-15) of FA in a polypropylene bottle. This polypropylene bottle was tightly closed, and it was heated at 308 K for 1 h. Then, this bottle was heated at 373 K for 6 h. The composite inside the bottle was transferred into a fused quartz reactor, which was equipped with a capillary cap to prevent air from entering from the outside while releasing gaseous by-products. The fused quartz reactor was heated to 1173 K
TE D
(heating rate of 3.5 K min-1), and the final temperature was maintained for 3 h. The reactor was cooled to room temperature, and the sample was dispersed in 20 mL of a 10 wt% hydrofluoric acid solution under magnetic stirring to remove the silica template. The resulting OMC was filtered, washed several times with distilled water, and dried at
EP
373 K.
2.1.3. OMC synthesis using mineral-acid-impregnated silica templates In a typical procedure for phosphoric acid impregnation in mesoporous silica, 0.1 g of calcined SBA-15 was
AC C
mixed with 5 mL of a 4 mM aqueous solution of phosphoric acid (85 wt%, Aldrich) in a beaker under magnetic stirring for 1 h. Then, the beaker was heated in a convection oven at 443 K for 12 h to evaporate water and anchor the
phosphoric acid onto the mesopore walls via covalent bond. The physically adsorbed phosphoric acid was removed by a repeated washing and filtration process using distilled water until the pH of the filtrate reaches 7. The product was dried at 373 K for 12 h. The Si/P ratio of the synthesized sample was 26. The impregnation of sulfuric acid was also performed following the same procedure as that employed in the phosphoric acid impregnation, except for the
use of sulfuric acid (49 wt%, Daejung) instead of the phosphoric acid. The SBA-15 impregnated with phosphoric acid
4
ACCEPTED MANUSCRIPT
or sulfuric acid is referred to as ‘PA-SBA-15’ and ‘SA-SBA-15’, respectively. The phosphorous and sulfur content in the silica template impregnated with mineral acids was determined by ICP-AES analysis. For OMC synthesis, we used various volumes of FA (equivalent to 130 or 170% of the total pore volume of the
RI PT
silica template). In a typical synthesis procedure, 0.1 g of PA-SBA-15 was added to a mixture of 2 ml of an ethanol solution and 0.13 ml (amounted to 130% of total pore volume of PA-SBA-15) of FA at room temperature. Then, the solution was magnetically stirred for 3 h and heated in an oil bath at 333 K under magnetic stirring to evaporate the ethanol solvent. The composite was placed in a drying oven at 373 K for 8 h. The FA polymerization, subsequent
SC
carbonization and template removal were performed according to the procedure used for OMC synthesis using Al-
SBA-15. The resulting OMC samples synthesized with FA was equivalent to 130% and 170% of template total pore
M AN U
volume are referred to as 1.3FA-OMC and 1.7FA-OMC, respectively.
2.2. Characterization
Powder X-ray diffraction (XRD) patterns were recorded on a Rigaku Multiplex diffractometer using a monochromatized X-ray beam from Cu Kα radiation (40 kV, 30 mA). The XRD scanning was performed under ambient conditions at step sizes of 0.01 with an accumulation time of 1.0 s. Nitrogen adsorptiondesorption isotherms were measured at 77 K using a volumetric gas sorption analyzer (Micromeritics Tristar
TE D
II). Prior to the measurement, the samples were degassed under vacuum for 3 h at 573 K. The specific surface area of the samples was calculated using the Brunauer-Emmett-Teller (BET) method with the adsorption data at a relative pressure range of 0.05 ~ 0.2. The pore size distribution was derived using the Barrett-Joyner-Halenda (BJH) method for isotherm analysis. A total pore volume was estimated from the
EP
amount of N2 adsorbed at P/P0=0.95. Transmission electron microscopy (TEM) images were recorded on a Tecnai G2 F30 and TitanTM ETEM G2 at an operating voltage of 300 kV. The X-ray photoelectron spectroscopy (XPS) experiments were performed with a Sigma Probe (Thermo VG Scientific) equipped with
AC C
a microfocused monochromator X-ray source. Temperature-programmed desorption (TPD) of NH3 was measured in a He flow using BELCAT-M
(BEL Japan) equipped with a thermal conductivity detector (TCD). Prior to the measurement, 50 mg of each
sample was placed in a quartz reactor and degassed for 3 h at 523 K. After cooling to 373 K, the NH3 gas was flowed into the reactor for 1 h at a flow rate of 30 mL min-1. Then, the free and weakly adsorbed NH3
was removed by flowing He for 1 h at 373 K. Next, the TPD profiles were measured using TCD with a continuous temperature increase (10 K min-1) under a He flow (30 mL min-1). Thermogravimetric analysis
5
ACCEPTED MANUSCRIPT
(TGA) was performed with TGA Q50 in a platinum crucible, under nitrogen atmosphere. The temperature was increased to 623 K at a ramp rate of 10 K min-1 and subsequently maintained for 60 min. Nuclear magnetic resonance (NMR) spectra were acquired in the solid state with magic angle spinning
RI PT
(MAS) using a Bruker Avance III HD 400WB spectrometer at room temperature. 31P MAS NMR spectra were recorded in a single-pulse sequence with a pulse width of 2 µs, a relaxation time of 5 s, and a spinning frequency of 12 kHz. An 85% H3PO4 aqueous solution was used as an external reference for the 31P NMR
chemical shift. 13C cross-polarization (CP) MAS NMR spectra were recorded with a contact time of 2 ms, a relaxation time of 5 s, and a spinning frequency of 10 kHz. The carbonyl carbon of glycine was used as an
M AN U
3. Results and discussion
SC
external chemical shift reference for the 13 C NMR.
3.1. OMC synthesis using the Al-SBA-15 template with excess FA
The OMC sample was synthesized using an Al-SBA-15 template by impregnating an excess amount of FA corresponding to 130% of the total pore volume (Fig. S1). As shown in Fig. 1A, the XRD pattern of the OMC product exhibits three reflection peaks at a 2θ of 0.95, 1.6, and 1.8o. The positions of the peaks were nearly same as those of the (10), (11) and (20) reflection peaks for the 2D hexagonal structure in the Al-SBA-15 template. This result
TE D
indicates that the ordered mesostructure of the carbon sample was well replicated from the silica template. However, the relative intensities between the (10) and (11) peaks for the OMC product were significantly different from those of the silica template. In particular, the intensity of the (10) peak was relatively small. This result indicates that the OMC might be composed of a mixture of CMK-3 and CMK-5 structures. It is important to note that both CMK-3 and
EP
CMK-5 could be synthesized using the Al-SBA-15 template. However, CMK-3 and CMK-5 have distinctly different XRD patterns. CMK-3, which is composed of nanorods with 2D hexagonal order, exhibits an intense (10) peak [31-
AC C
33]. On the other hand, CMK-5, which consists of a hollow nanopipe framework, exhibits negligible intensity for the (10) reflection peak due to ‘accidental extinction’ [34]. A TEM investigation confirmed that the OMC sample was composed of a mixture of nanorod-type CMK-3 and nanopipe-type CMK-5 structure (Figs. 1B and 1C). These structures were arranged with long-range hexagonal order over the entire region of the sample. However, as shown in Fig. 1D, the OMC product also has non-templated carbon species outside of the ordered mesoporous carbon
frameworks. When an excess amount of FA was impregnated into the Al-SBA-15 template, a certain amount of FA remained outside of the template. The remaining FA molecules polymerized rapidly due to the exothermic polymerization catalyzed by the strong acid sites in Al-SBA-15. The polymerized FA outside the template carbonized into non-
6
ACCEPTED MANUSCRIPT
templated carbon species. In addition to the extra FA, FA molecules, which were supposed to fill the template mesopores, could also be polymerized outside before they homogenously infiltrated into the mesopores due to the heat generated from the exothermic polymerization. A mixture of nanorods and nanopipes was obtained as the OMC
RI PT
product. From this result, it can be inferred that the selective carbon replication without external carbon species is not possible with Al-SBA-15 template via one-step impregnation. To improve the problem, the moderately weak acidity
of the template is required to provide sufficient time for the infiltration of FA into the template mesopores before the
purpose, we introduce a mineral acid impregnation method.
3.2. OMC synthesis using the PA-SBA-15 template with excess FA
SC
polymerization, and to prevent the generation of non-templated carbons during the carbonization process. For this
M AN U
We attempted to use PA-SBA-15 as an alternative method for synthesizing OMC materials. The PA-SBA-15 sample was prepared by phosphoric acid impregnation. The results of the phosphoric acid impregnation into SBA-15 are summarized in Fig. 2 and Fig. S1 in the Supplementary data. As shown in Fig. 2A, the XRD pattern of PA-SBA15 reveals that the ordered hexagonal (p6mm) mesostructure of SBA-15 was fully retained after the phosphoric acid functionalization. The N2 adsorption-desorption isotherm of PA-SBA-15 exhibits a similar appearance to that of pristine SBA-15. However, a stiff increase in the range of P/P0 between 0.6 and 0.8 was slightly shifted to lower P/P0
TE D
compared to the SBA-15 sample. The pore size distribution was determined from the adsorption branch of the isotherm by employing the BJH method. The mesopore size distribution of PA-SBA-15 is centered at 8.8 nm, which is slightly smaller than that of SBA-15 (9.1 nm) (Fig. S1B). The specific BET surface area and total pore volume of PA-SBA-15 (570 m2 g-1 and 1.00 cm3 g-1, respectively) are comparable to those of SBA-15 (690 m2 g-1 and 1.14 cm3
EP
g-1, respectively). The N2 adsorption isotherm results indicate that phosphoric acid was most likely supported on the mesopore walls of SBA-15 without pore blocking. The chemical state of the phosphate groups on the mesopore walls of PA-SBA-15 was characterized using solid
AC C
state 31P NMR spectroscopy. The 31P NMR spectrum of PA-SBA-15 contains three resonance peaks at 0.2, -11, and 23 ppm as shown in Fig. 2B. Among these peaks, the peak at 0.2 ppm was assigned to the phosphoric acids that were physically adsorbed on the mesopore walls via hydrogen bonds [35]. The peaks at -11 and -23 ppm correspond to the phosphate ester and phosphate diester, respectively, which are covalently bound to the mesopore walls [35, 36]. The
deconvolution and peak quantification of the NMR spectrum were performed to measure the amount of each species in PA-SBA-15. The molar ratio of the hydrogen-bonded phosphoric acid, phosphate ester and phosphate diester was
1.0: 10: 3.1. This result indicates that most of the phosphoric acid species were covalently bound to the silanol groups in the SBA-15 framework and formed phosphate ester groups. These species have weak Brönsted acidity [37]. The
7
ACCEPTED MANUSCRIPT
acidity of PA-SBA-15 was measured by the desorption peak in the NH3 TPD profile below 600 K (Fig. S2). This desorption peak was attributed to NH3 desorption from the Brönsted and Lewis acid sites in the framework. The peak position of PA-SBA-15 is much lower than that of Al-SBA-15. This result indicates that the strength of the acid sites
RI PT
in PA-SBA-15 are lower than those in Al-SBA-15. Using PA-SBA-15 as a template, 1.3FA-OMC was synthesized with the same amount of FA used for the AlSBA-15-templated OMC sample. The small-angle XRD pattern of the 1.3FA-OMC sample exhibits three Bragg
reflection peaks that can be indexed to the (10), (11), and (20) reflections of the 2D hexagonal structure (Fig. 3A). In
SC
contrast to the XRD pattern of Al-SBA-15-templated OMC, the intensity of the (10) reflection peak was significantly large. Therefore, the relative intensity ratio between (10) and (11) is similar to that of typical CMK-3 [32, 33]. This result indicates that 1.3FA-OMC would have a rod-type CMK-3 structure. The N2 adsorption-desorption isotherm of
M AN U
the 1.3FA-OMC sample exhibits a stiff increase in the range of P/P0 between 0.4 and 0.6, which indicates the presence of mesopores with a very uniform pore diameter (Fig. 3B). 1.3FA-OMC exhibits a narrow unimodal pore size distribution at 3.6 nm (Fig. 3B). The BET surface area and total pore volume of 1.3FA-OMC are very large (990 m2 g-1 and 0.90 cm3 g-1, respectively). The representative TEM image confirms that 1.3FA-OMC has a highly ordered hexagonal mesostructure, which is the exact inverse of the PA-SBA-15 template (Figs. 3C and D). Low magnification TEM images of 1.3FA-OMC confirmed that it was composed of nanorods without external carbon
TE D
species (Fig. S3).
It was also possible to synthesize OMC even when the amount of FA impregnated corresponded to 170% of the total pore volume of PA-SBA-15. The XRD and TEM investigation results assured that the 1.7FA-OMC product exhibited a well-ordered CMK-3 structure without non-templated carbons (Fig. S4). The mesopore diameter is very
EP
uniform with the size of 3.6 nm, which is the same as that of 1.3FA-OMC (Fig. S4C). The total pore volume and BET surface area of 1.7FA-OMC (940 m2 g-1, 0.78 cm3 g-1, respectively) were also comparable to those of CMK-3. It is important to note that faithful carbon replication was realized even when an excessive amount of FA was
AC C
impregnated within a single step. In comparison to Al-SBA-15, the acidity of PA-SBA-15 was considerably weak. The weak acidity restricted the rapid polymerization of FA molecules, which allowed the homogeneous infiltration of FA into the template mesopores. Meanwhile, the extra FA was only polymerized to a low degree, and this material was eventually vaporized during the high-temperature step, which prevented the generation of external carbon (This
will be discussed below). Therefore, a faithful CMK-3 carbon replica was obtained without external carbon species. Interestingly, the current method can result in the incorporation of P atoms in the framework of OMC. The solid
state 31P NMR spectrum of 1.3FA-OMC exhibits two resonance peaks at -6.2 and 8.5 ppm as shown in Fig. S5A. These peaks can be assigned to the PPh3 and RPh2 groups, respectively, which are most likely generated in the carbon
8
ACCEPTED MANUSCRIPT
framework. During the carbonization process, phosphate esters on the mesopore walls of the silica template are covalently bound to the carbon framework. Then, only the silica species were removed by HF while the P atoms remained in the carbon frameworks as PPh3 and RPPh2 species. The presence of P atoms in the carbon framework
RI PT
was also confirmed by XPS analysis. The XPS spectrum of 1.3FA-OMC contained two peaks centered at binding energies of 132.9 and 138 eV, which correspond to P-C and P-O bonds, respectively (Fig. S5B). From the XPS result, the content of P atoms in 1.3FA-OMC was determined to be 2.0 wt%. Such incorporation of heteroatoms into the
redox catalytic reactions [16, 38, 39].
3.3. NMR characterization of the carbonization process
SC
carbon framework can alter the intrinsic chemical properties and therefore can result in catalytic function in various
M AN U
The polymerization and carbonization process of 1.3FA-OMC were monitored by solid state 13C CP-MAS NMR. Fig. 4A shows the NMR spectra of the carbon-silica composites that were collected after thermal treatment at various temperatures. In addition, the liquid NMR spectrum of the FA monomer is shown in Fig. 4A. The NMR spectrum of the FA monomer contains five distinct peaks. Among these peaks, the peak at 56 ppm was assigned to the hydroxymethyl carbon atom (designated ‘a’ in the molecular structure of FA, shown in Fig. 4B) of FA [40]. Other four peaks in the range of 100~160 ppm correspond to the carbon atoms in the furan ring of FA.
TE D
All of the peaks from FA were broadened when the FA-containing silica was thermally treated at 333 K. The peak broadening indicates that the FA molecules were adsorbed in mesopores and started to convert into polymeric species. As the temperature increased to 373 K, the peak for the ‘a’ carbon of FA nearly disappeared, and the other peaks corresponding to the furan ring of FA became much broader. In addition, several new peaks in the range of
EP
10~40 ppm were observed. Fig. 4B shows that such a dramatic change in the NMR spectrum was due to the polymerization of FA molecules in the mesoporous silica template. During the polymerization step, the peak corresponding to the ‘a’ carbon disappeared as the alcohol groups were condensed to form C-C bonds between the
AC C
FA molecules. Instead, terminated carbons or methylene bridges between the furan rings were generated, which were confirmed by the peaks in the range of 10 ~ 40 ppm [40]. It should be noted that the rate of the FA polymerization in PA-SBA-15 is clearly different from that in Al-SBA-15 (Fig. S6). The NMR spectrum of FA-containing Al-SBA-15
obtained at 333 K is nearly same as that of PA-SBA-15 at 373 K. This result indicates that the FA molecules in the
Al-SBA-15 sample were very rapidly polymerized even at 333 K. In contrast to Al-SBA-15, the weak acid sites in PA-SBA-15 allowed the slow polymerization of FA and thus the homogenous infiltration of FA into the template. When the composite sample was thermally treated at 623 K, a broad peak at 124 ppm appeared in the NMR spectrum. This peak indicates that the polymeric FA species were transformed into aromatic carbon frameworks. In
9
ACCEPTED MANUSCRIPT
addition, there are several sharp peaks in the range of 0 ~ 50 ppm that could be assigned to alkane hydrocarbons. These hydrocarbons were likely generated from the thermal decomposition of low-degree polymerized FA outside the template. In the NMR spectrum of the composite sample collected at 1173 K, all of the peaks corresponding to
RI PT
the carbon atoms in the polymeric FAs and alkane hydrocarbons disappeared. Only a broad peak at 122 ppm corresponding to the aromatic carbon framework was observed. Based on this NMR analysis result, it can be inferred
that the FA in the PA-SBA-15 template polymerized slowly at an early stage. Therefore, there was sufficient time for homogeneous infiltration into the template mesopores. While the infiltrated FA in PA-SBA-15 were gradually
SC
polymerized and carbonized, the less polymerized extra FA outside the template decomposed and vaporized. To
confirm this assumption, we performed TGA experiments of both polymerized FA containing PA-SBA-15 and AlSBA-15 samples (Fig. S7). The FA was polymerized at 373 K. Both of the samples showed the weight loss between
M AN U
373 K and 623 K, which indicates the decomposition of polymerized FA. However, the weight loss of PA-SBA-15 composite was much larger than that of Al-SBA-15 composite. This means that the degree of the polymerization in PA-SBA-15 was lower than that in Al-SBA-15. Considering that the carbon replica obtained from PA-SBA-15 was composed of rigid carbon nanorods, the less polymerized FA species were located outside of the pores. Therefore, a faithful carbon replica was obtained using the PA-SBA-15 template regardless of the amount of FA infiltrated into
TE D
the template.
3.4. OMC Synthesis using different silica templates and mineral acids The OMC synthesis method using a phosphoric acid impregnated silica template was applied to the synthesis of OMC with various mesoporous silicas. Two different mesoporous silicas SBA-16 (cubic Im3m) and KIT-6 (cubic
EP
Ia3d) were used as templates for carbon replication. The phosphate functionalization of the silica template and carbon replication process was performed following the same procedure used for PA-SBA-15. As shown in Fig. 5A, the resulting carbon replica of phosphoric acid impregnated KIT-6 exhibits well-resolved peaks in the XRD pattern.
AC C
These peaks can be indexed to the (211), (220), and (420) reflections of the cubic Ia3d structure and the peak position is nearly same as that of the KIT-6 template [41]. The representative TEM image of the carbon replica shows a highly ordered structure (Fig. 5B). In addition, the shape of the N2 adsorption-desorption isotherm shows a typical type IV isotherm, which indicates pore condensation with hysteresis (Fig. S8). The mesopore diameter of the carbon replica was 3.8 nm, and the BET surface area and total pore volume were calculated to be 910 m2 g-1 and 1.0 cm3 g-1,
respectively. The XRD pattern for the carbon replica of phosphoric acid impregnated SBA-16 is characteristic of the cubic Im3m space group (Fig. 5C) and similar to the XRD pattern reported by Kim et al [6]. The TEM image of the carbon replica confirms that the carbon replica has a structural ordering with cubic symmetry (Fig. 5D). The BET
10
ACCEPTED MANUSCRIPT
surface area and total pore volume (780 m2 g-1 and 0.7 cm3 g-1, respectively) of the carbon replica are also comparable to that of CMK-6 [6]. Based on the results obtained from the phosphoric acid impregnation method, it was possible to extend the
RI PT
current method to different mineral acids, such as sulfuric acid. Sulfuric-acid modified SBA-15 (SA-SBA-15) was prepared by the same procedure used in PA-SBA-15, except that sulfuric acid was used instead of phosphoric acid.
The acidity of SA-SBA-15 was measured by NH3 TPD (Fig. S9). The peak of NH3-desorption from the acid sites was only observed in the low temperature region (< 500 K), which indicates that there are no strong acid sites. This result
SC
indicates that the sulfuric acid might be functionalized on the surface of SA-SBA-15 in the form of sulfate esters
which exhibit weak acidity. The small-angle XRD pattern of the OMC product synthesized with the SA-SBA-15 template exhibits three well-resolved Bragg reflection peaks that are similar to those of 1.3FA-OMC (Fig. 6A). The
M AN U
representative TEM image of the OMC product also exhibits a highly ordered hexagonal mesostructure (Fig. 6B). While the pore size distribution curve shows a sharp peak centered at 3.6 nm (Fig. S10), the BET surface area and total pore volume is 950 m2 g-1 and 0.86 cm3 g-1, respectively. In contrast to the phosphoric acid impregnation method, the elementary analysis result revealed that there are no heteroatoms in the framework. In comparison to the phosphorus, the sulfur could be easily decomposed and vaporized during the high-temperature carbonization process.
4. Conclusion
TE D
Therefore, the OMC with a pure carbon framework was obtained when SA-SBA-15 was used as a template.
We demonstrated a simple method for CMK-3 synthesis using phosphoric acid-impregnated SBA-15 as a template. The 31P NMR results indicated that impregnated phosphoric acids were functionalized by forming
EP
phosphate ester groups on the mesopore walls of SBA-15. The mild acidity of the phosphate ester in the SBA-15 template allowed the slow polymerization of FA, which prevented the generation of external carbon species even
AC C
when a large amount of FA was infiltrated. Therefore, the phosphoric acid-impregnated SBA-15 led to selective carbonization inside the mesopores and yielded highly faithful carbon replicas. The sulfuric acid could also be applied to the acid-functionalization of SBA-15 for the CMK-3 synthesis. Furthermore, this method using mineral acid impregnation could be extended to various silica templates such as KIT-6 and SBA-16. We expect that the mineral acid impregnation method would be useful for the carbon replication of various kinds of silica templates such as nanotubes, nanospheres, pure siliceous zeolites, and hierarchically porous materials.
11
ACCEPTED MANUSCRIPT
Acknowledgments This work was supported by IBS-R004-D1.
RI PT
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/xxx
[1] R. Ryoo, S. H. Joo, S. Jun, J. Phys. Chem. B 103 (1999) 7435.
SC
References
[2] S. Jun, S. H. Joo, R. Ryoo, M. Kruk, M. Jaroniec, Z. Liu, T. Ohsuna, O. Terasaki, J. Am. Chem. Soc. 122
M AN U
(2000) 10712.
[3] S. H. Joo, S. J. Choi, I. Oh, J. Kwak, Z. Liu, O. Terasaki, R. Ryoo, Nature 412 (2001) 169. [4] R. Ryoo, S. H. Joo, M. Kruck, M. Jaroniec, Adv. Mater. 13 (2001) 677. [5] J.-S. Lee, S. H. Joo, R. Ryoo, J. Am. Chem. Soc. 124 (2002) 1156.
[6] T.-W. Kim, R. Ryoo, K. P. Gierszal, M. Jaroniec, L. A. Solovyov, Y. Sakamoto, O. Terasaki, J. Mater. Chem. 15 (2005) 1560.
[7] C. Liang, K. Hong, G. A. Guiochon, J. W. Mays, S. Dai, Angew. Chem. Int. Ed. 43 (2004) 5785.
TE D
[8] F. Zhang, Y. Meng, D. Gu, Y. Yan, C. Yu, B. Tu, D. Zhao J. Am. Chem. Soc. 127 (2005) 13508. [9] M. Choi, R. Ryoo, J. Mater. Chem. 17 (2007) 4204.
[10] K.-I. Min, J.-S. Choi, Y.-M. Chung, W.-S. Ahn, R. Ryoo, P. K. Lim Appl. Cat. A: General 337 (2008) 97. [11] H. Zhou, S. Zhu, M. Hibino, I. Honma, M. Ichihara, Adv. Mater. 15 (2003) 2107.
EP
[12] N.-I. Kim, J. Y. Cheon, J. H. Kim, J. Seong, J.-Y. Park, S. H. Joo, K. Kwon, Carbon 72 (2014) 354. [13] A. Vinu, P. Srinivasu, M. Takahashi, V. V. Balasubramanian, K. Ariga, Micropor. Mesopor. Mater. 153
AC C
(2001) 44.
[14] H. I. Lee, J. H. Kim, D. J. You, J. E. Lee, J. M. Kim, W.-S. Ahn, C. Pak, S. H. Joo, H. Chang, D. Seung, Adv. Mater. 20 (2008), 757.
[15] J. Kim, M. Choi, R. Ryoo, Bull. Korean. Chem. Soc. 29 (2008) 413. [16] D.-S. Yang, D. Bhattacharjya, S. Inamdar, J. Park, J.-S. Yu, J. Am. Chem. Soc. 134 (2012) 16127. [17] X. Zhao, Q. Zhang, B. Zhang, C.-M. Chen, A. Wang, T. Zhang, D. S. Su, J. Mater. Chem. 22 (2012) 4963.
[18] D. Zhang, Y. Hao, L. Zheng, Y. Ma, H. Feng, H. Luo, J. Mater. Chem. A 1 (2013) 7584. [19] T.-W. Kim, I.-S. Park, R. Ryoo, Angew. Chem. Int. Ed. 42 (2003) 4375.
12
ACCEPTED MANUSCRIPT
[20] J. Lee, S. Yoon, T. Hyeon, S. M. Oh, K. B. Kim, Chem. Commun. (1999) 2177. [21] O. J. A. Schueller, S. T. Brittain, G. M. Whitesides, Adv. Mater. 9 (1997) 477. [22] G. Tondi, A. Pizzi, R. Olives Maderas. Ciencia y tecnologia 10 (2008) 219.
RI PT
[23] T.-W. Kim, L. A. Solovyov, J. Mater. Chem. 16 (2006) 1445. [24] K. P. Gierszal, M. Jaroniec, T.-W. Kim, J. Kim, R. Ryoo, New J. Chem. 32 (2008) 981. [25] Z. Tian, M. A. Snyder, Langmuir 30 (2014) 12411. [26] Z. Tian, M. A. Snyder, Langmuir 30 (2014) 9828.
SC
[27] A.-H. Lu, W.-C. Li, W. Schmidt, W. Kiefer, F. Schüth, Carbon 42 (2004) 2939.
[28] P. A. Bazula, A.-H. Lu, J.-J. Nitz, F. Schüth, Micropor. Mesopor. Mater. 108 (2008) 266.
[29] A. Galarneau, H. Cambon, F. D. Renzo, R. Ryoo, M. Choi, F. Fajula, New J. Chem. 27 (2003) 73.
M AN U
[30] T.-W. Kim, F. Kleitz, B. Paul, R. Ryoo, J. Am. Chem. Soc. 127 (2005) 7601. [31] I.-S. Park, M. Choi, T.-W. Kim, R. Ryoo, J. Mater. Chem. 16 (2006), 3409. [32] A. B. Fuertes, Micropor. Mesopor. Mater. 67 (2004) 273.
[33] J. Roggenbuck, G. Koch, M. Tiemann, Chem. Mater. 18 (2006) 4151.
[34] K. Lund, N. Muroyama, O. Terasaki, Micropor. Mesopor. Mater. 128 (2010) 71. [35] L.-Q. Wang, W. D. Samuels, G. J. Exarhos, B. I. Lee, Z. Cao, J. Mater. Chem. 8 (1998) 165.
TE D
[36] G. D. Cody, B. Mysen, G. Sághi-Szabó, J. A. Tossell, Geo. et Cos. Acta 65 (2001) 2395. [37] T. L. Lamoureaux, D. H. Lee, Chem. Commun. 47 (2011) 8623. [38] J. Wu, Z. Yang, X. Li, Q. Sun, C. Jin, P. Strasser, R. Yang, J. Mater. Chem. A 1 (2013) 9889. [39] P. Karthika, N. Rajalakshmi, K. S. Dhathathreyan, J. Nanosci. Nanotech. 13 (2013) 1746.
EP
[40] L.-T. Cheng, W. J. Tseng, J. Polym. Res. 17 (2010) 391.
AC C
[41] F. Kleitz, S. H. Choi, R. Ryoo, Chem. Commun. (2003) 2136.
13
ACCEPTED MANUSCRIPT
M AN U
SC
RI PT
Figures
Fig. 1. (A) XRD pattern and (B-D) TEM images of the OMC product obtained using the Al-SBA-15 template. (B) and (C) are nanorods and nanopipes, respectively, in the OMC product. White arrows in (D) indicate external carbon
AC C
EP
TE D
species. For comparison, the XRD pattern of Al-SBA-15 is included in (A).
Fig. 2. (A) Powder XRD pattern and (B) Solid state 31P MAS NMR spectrum of PA-SBA-15. For comparison, the XRD pattern of SBA-15 is included in (A).
14
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Fig. 3. (A) Powder XRD pattern, (B) N2 adsorption-desorption isotherm and mesopore size distribution corresponding to the adsorption branch, and (C) TEM image of 1.3-FA-OMC. For comparison, a TEM image of PA-
AC C
EP
SBA-15 is shown in (D).
15
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Fig. 4. (A) Liquid state 13C NMR spectrum of the FA monomer and solid state 13C CP-MAS NMR spectra of the
AC C
EP
TE D
carbon-silica composite samples collected at various temperatures and (B) scheme for FA polymerization.
16
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Fig. 5. Powder XRD patterns and TEM images of carbon replica of (A, B) KIT-6 and (C, D) SBA-16. For
AC C
EP
comparison, the XRD pattern of the KIT-6 and SBA-16 template was included.
Fig. 6. (A) Powder XRD pattern and (B) TEM image of OMC obtained from Sulfuric-acid modified-SBA15.
17
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
18
ACCEPTED MANUSCRIPT
Highlights
AC C
EP
TE D
M AN U
SC
RI PT
Mineral acid could be simply anchored on silica templates through ester bonding. > The functionalized silicas showed moderate acidity for faithful carbon replication. > The moderate acidity led to selective carbon formation inside the template pores. > External carbon deposition was prevented despite using carbon precursor excessively. > The present method could be extended for various mesoporous silica templates.
ACCEPTED MANUSCRIPT
Synthesis of Mesoporous Carbons using Silica Templates
RI PT
Impregnated with Mineral Acids
Yongbeom Seo,† Kyoungsoo Kim,‡ Younjae Jung,† and Ryong Ryoo‡,†,*
Department of Chemistry, Korea Advanced Institute of Science and Technology (KAIST),
SC
†
Daejeon 305-701, Korea
Center for Nanomaterials and Chemical Reactions, Institute for Basic Science (IBS), Daejeon
M AN U
‡
305-701, Korea
*Corresponding author: Ryong Ryoo (E-mail address: [email protected]; Tel.: +82 42 350
AC C
EP
TE D
2830)
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
TE D
Figure S1. (A) N2 isotherm and (B) pore size distribution of SBA-15 and PA-SBA-15. Isotherms and pore size distributions of samples are vertically offset by 400 cm3 g-1 and 5.0 cm3 g-1, respectively. The BET surface area and total pore volume of Al-SBA-15 are
AC C
EP
625 m2 g-1 and 1.13 cm3 g-1, respectively.
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Figure S2. NH3 TPD profiles of PA-SBA-15 and Al-SBA-15. The result for PA-SBA15 shows an additional peaks above 800 K, but it is caused by thermal decomposition of
EP
phosphate esters and condensation of silanol groups. For comparison, TPD profile of
AC C
PA-SBA-15 without NH3 is included.
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
Figure S3. Low-magnification TEM images of 1.3FA-OMC.
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
EP
Figure S4. (A) Powder XRD pattern, (B) TEM image, and (C) N2 adsorptiondesorption isotherm and mesopore size distribution corresponding to the adsorption
AC C
branch of 1.7FA-OMC.
31
M AN U
Figure S5. (A) Solid state
SC
RI PT
ACCEPTED MANUSCRIPT
P MAS NMR spectrum and (B) XPS spectrum of 1.3FA-
OMC. ‘*’ in (A) indicates spinning side bands. The red and blue curves in (B) indicate
AC C
EP
TE D
results of spectral analysis by Gaussian deconvolution.
Figure S6. Solid state
13
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
C CP-MAS NMR spectrum of carbon/Al-SBA-15 composite
AC C
EP
TE D
sample collected at 333 K.
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
Figure S7. TGA curves for the polymerized FA-containing silica templates. The temperature was increased to 623 K in N2 flow at a ramp rate of 10 K min-1 and subsequently maintained for 60 min.
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Figure S8. N2 isotherm and pore size distribution of SBA-16 and KIT-6 template and
AC C
their carbon replicas. Isotherms and pore size distributions of samples are vertically offset by 100 cm3 g-1 and 1.5 cm3 g-1, respectively.
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Figure S9. NH3 TPD profile of SA-SBA-15. The result for SA-SBA-15 shows an
TE D
additional peaks above 800 K, but it is caused by condensation of silanol groups and
AC C
EP
thermal decomposition of sulfate ester groups.
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Figure S10. N2 isotherm and pore size distribution of carbon replica of SA-SBA-15.
AC C
EP
TE D
.
S1387-1811(15)00033-5
DOI:
10.1016/j.micromeso.2015.01.021
Reference:
MICMAT 6946
To appear in:
Microporous and Mesoporous Materials
Received Date: 2 October 2014 Revised Date:
7 January 2015
Accepted Date: 15 January 2015
Please cite this article as: Y. Seo, K. Kim, Y. Jung, R. Ryoo, Synthesis of Mesoporous Carbons using Silica Templates Impregnated with Mineral Acids, Microporous and Mesoporous Materials (2015), doi: 10.1016/j.micromeso.2015.01.021. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Synthesis of Mesoporous Carbons using Silica Templates
RI PT
Impregnated with Mineral Acids Yongbeom Seo,† Kyoungsoo Kim,‡ Younjae Jung,† and Ryong Ryoo‡,†,*
Department of Chemistry, Korea Advanced Institute of Science and Technology (KAIST), Daejeon 305-701, Korea
‡
Center for Nanomaterials and Chemical Reactions, Institute for Basic Science (IBS), Daejeon 305-701, Korea
*
Corresponding author: Ryong Ryoo (E-mail: [email protected]; Tel.: +82 42 350 2830)
M AN U
SC
†
Abstract
The pore walls of SBA-15, SBA-16 and KIT-6 mesoporous silica were acidified via the impregnation of phosphoric acid or sulfuric acid prior to their use as a template for mesoporous carbon synthesis with furfuryl alcohol. The effect of acid functionalization on carbon synthesis was investigated using X-ray powder diffraction, N2 adsorption, transmission electron microscopy, and solid-state 13C and 31P NMR spectroscopy. The results indicated that a
TE D
significant portion of the impregnated mineral acid formed an ester with the surface silanol group. Therefore, the impregnated acid could lead to the selective formation of carbon frameworks inside the template pores, preventing external carbon deposition even when furfuryl alcohol was used excessively. The mineral acid impregnation yielded
EP
highly faithful carbon replicas of the mesoporous silica.
AC C
Keywords: template synthesis; mesoporous material; ordered mesoporous carbon; mineral acid; solidstate NMR
1
ACCEPTED MANUSCRIPT
1. Introduction Mesopores are defined as pores in the range of 2 to 50 nm in diameter. Carbon materials with regular arrays of these mesopores are called ordered mesoporous carbons (OMCs) [1-8]. OMCs with various structures are currently
RI PT
available via a synthetic route using mesoporous silica templates [1-6] or carbonization of ordered arrays of
polymeric resin nanobeads [7, 8]. In the current work, we focus on the template synthesis. The template synthesis of
OMCs was first reported by Ryoo et al. in 1999 [1]. These types of OMCs are referred to as ‘CMK’. The CMK-type of carbon materials have attracted much attention as absorbents, catalyst supports, and electrode materials [3, 9-12].
SC
There have been numerous reports on the synthesis, characterization and application of CMK carbons with various
structures and pore diameters [2-6, 9-14]. Surface modification and functionalization of CMK carbons have also been extensively studied to improve their performance [15-18].
M AN U
The synthesis procedure for CMK carbons involves three steps [1-5]. In the first step, an organic compound is impregnated into the mesopores of a silica template. The impregnated organic source is polymerized and subsequently converted to carbon by pyrolysis. The key in this step is to control the carbonization process to occur inside the template mesopores. The confinement of the carbonization process within the mesopores can be realized by converting the organic source to cross-linked solid polymers prior to the final conversion to carbon. The polymerization can be achieved by the incorporation of an acid into the silica template or the carbon source. The
TE D
silica template can be removed after carbonization with hydrofluoric acid or a sodium hydroxide solution. Ultimately, the carbon product should become a faithful inverse replica of the porous silica template. Various organic compounds, such as sucrose [1, 2, 5], furfuryl alcohol (FA) [3], acenaphthene [19], phenol resin [20], and acrylonitrile [15], can be used as the carbon source. FA is one of the most convenient sources of carbon as
EP
the liquid-phase compound that can be easily impregnated into silica mesopores. Nevertheless, the carbonization process using FA involves several time consuming steps. For example, incorporation of a solid acid (typically Al) onto the template pore walls is necessary. The procedure involves wet impregnation of an Al compound and
AC C
evaporation of the solvent followed by calcination. Then, FA is impregnated using an incipient wetness process. The
impregnated FA is polymerized by heating at 373 K due to the catalytic function of the acid sites. Upon further heating to 623 K, the carbon source is converted to a black tar with a significant contraction in the volume. An
additional amount of FA can be added to compensate for the volume loss (i.e., second impregnation) at this point.
Then, the procedures for FA polymerization at 373 K and subsequent tar formation are repeated. After all of these treatments, the silica template containing the carbon source is heated to the final carbonization temperature (~ 1173 K). The resultant carbon from the two-step FA impregnation exhibits a rod-type framework that fills the template pores. If the second FA impregnation is skipped, the carbon often becomes a hollow pipe-type framework [3].
2
ACCEPTED MANUSCRIPT
It is important not to exceed the amount of FA beyond the total pore volume in a template. An excessive amount of FA can cause the formation of non-templated (i.e., external) carbon. The acid strength of the template pore walls is also an important factor. Often the polymerization rate of carbon sources including FA, sucrose, glucose, and other
RI PT
organics are very sensitive to acid catalysts [21 - 27]. For the aluminosilicate template, the acid strength is often so high that FA polymerization proceeds very rapidly before a sufficient time can allow for homogeneous infiltration of
FA into the template pores. The rapid polymerization can be a serious problem particularly in a large batch synthesis. The exothermic polymerization process can generate substantial heat that accelerates the polymerization itself [21,
SC
22]. The FA polymerization outside the template can result in the formation of external carbon. There were
alternative routes involving the use of silica templates without Al incorporation [23 - 27]. In such cases, a small amount of p-toluene sulfonic acid or oxalic acid were added to the FA as a polymerization catalyst prior to the FA
M AN U
being impregnated. This route is prone to suffering from the generation of polymerization heat as well as the formation of external carbon.
The current work was undertaken to improve the synthesis procedure for OMC materials using FA in mesoporous silica templates. In particular, this study was aimed at preventing the generation of external carbon by selecting a moderately weak acid that would slowly catalyze the FA polymerization. As a first candidate, phosphoric acid was impregnated into ordered mesoporous silica prior to FA infiltration. The resulting carbonization process was
TE D
investigated using X-ray powder diffraction, N2 adsorption, and transmission electron microscopy. The chemical state of the phosphoric acid in the silica template was analyzed by solid-state 31P NMR spectroscopy. Based on the chemical information from the 31P NMR data, we were able to extend the phosphoric acid-based method to the use of sulfuric acid. In this paper, we introduce a mineral acid impregnation method, and its application to the synthesis of
EP
OMCs with various mesoporous silica templates, such as SBA-15, KIT-6 and SBA-16.
2. Experimental
AC C
2.1. Synthesis
2.1.1. Synthesis of silica templates SBA-15 silica was synthesized according to the procedure described in the literature [29] using tetraethoxysilane
(TEOS, TCI) as a silica source and Pluronic P123 [EO20PO70EO20 (EO = ethylene oxide, PO = propylene oxide, Mn ~5800, Aldrich] as the mesopore-directing surfactant. The gel composition was 1 TEOS: 0.017 Pluronic P123: 5.3 HCl: 194 H2O. All of the starting materials were mixed at 308 K for 12 h using a magnetic stirrer. Then, the resulting
gel mixture was heated for 12 h at 403 K under static conditions in a Teflon-lined autoclave. The precipitated SBA-
3
ACCEPTED MANUSCRIPT
15 product was filtered, washed with deionized water and dried at 373 K for 12 h. The product was calcined in air at 823 K for 4 h. SBA-16 and KIT-6 were synthesized according to previously reported procedures [6, 30].
RI PT
2.1.2. OMC synthesis using aluminated SBA-15 In a typical procedure for post-synthetic alumination of SBA-15 with Si/Al of 20, 0.5 g of the calcined SBA-15
silica was added to 20 mL of ethanol (99.9%, Merck) containing 0.055 g of AlCl3 (anhydrous, Aldrich). The mixture was magnetically stirred at room temperature for 6 h. Then, ethanol was removed using a rotary evaporator. The
SC
SBA-15 containing Al (i.e., Al-SBA-15) was further dried in an oven at 373 K for 3 h followed by calcination in an air stream at 823 K for 4 h. The inductively coupled plasma-atomic emission spectroscopy (ICP-AES) analysis indicated that the Si/Al ratio of Al-SBA-15 was 22.
M AN U
An ordered mesoporous carbon (OMC) sample was prepared using Al-SBA-15 as a template and FA (98%, Aldrich) as a carbon precursor. In a typical synthesis, 0.1 g of Al-SBA-15 was impregnated with 0.14 mL (amounting to 130% of the total pore volume of Al-SBA-15) of FA in a polypropylene bottle. This polypropylene bottle was tightly closed, and it was heated at 308 K for 1 h. Then, this bottle was heated at 373 K for 6 h. The composite inside the bottle was transferred into a fused quartz reactor, which was equipped with a capillary cap to prevent air from entering from the outside while releasing gaseous by-products. The fused quartz reactor was heated to 1173 K
TE D
(heating rate of 3.5 K min-1), and the final temperature was maintained for 3 h. The reactor was cooled to room temperature, and the sample was dispersed in 20 mL of a 10 wt% hydrofluoric acid solution under magnetic stirring to remove the silica template. The resulting OMC was filtered, washed several times with distilled water, and dried at
EP
373 K.
2.1.3. OMC synthesis using mineral-acid-impregnated silica templates In a typical procedure for phosphoric acid impregnation in mesoporous silica, 0.1 g of calcined SBA-15 was
AC C
mixed with 5 mL of a 4 mM aqueous solution of phosphoric acid (85 wt%, Aldrich) in a beaker under magnetic stirring for 1 h. Then, the beaker was heated in a convection oven at 443 K for 12 h to evaporate water and anchor the
phosphoric acid onto the mesopore walls via covalent bond. The physically adsorbed phosphoric acid was removed by a repeated washing and filtration process using distilled water until the pH of the filtrate reaches 7. The product was dried at 373 K for 12 h. The Si/P ratio of the synthesized sample was 26. The impregnation of sulfuric acid was also performed following the same procedure as that employed in the phosphoric acid impregnation, except for the
use of sulfuric acid (49 wt%, Daejung) instead of the phosphoric acid. The SBA-15 impregnated with phosphoric acid
4
ACCEPTED MANUSCRIPT
or sulfuric acid is referred to as ‘PA-SBA-15’ and ‘SA-SBA-15’, respectively. The phosphorous and sulfur content in the silica template impregnated with mineral acids was determined by ICP-AES analysis. For OMC synthesis, we used various volumes of FA (equivalent to 130 or 170% of the total pore volume of the
RI PT
silica template). In a typical synthesis procedure, 0.1 g of PA-SBA-15 was added to a mixture of 2 ml of an ethanol solution and 0.13 ml (amounted to 130% of total pore volume of PA-SBA-15) of FA at room temperature. Then, the solution was magnetically stirred for 3 h and heated in an oil bath at 333 K under magnetic stirring to evaporate the ethanol solvent. The composite was placed in a drying oven at 373 K for 8 h. The FA polymerization, subsequent
SC
carbonization and template removal were performed according to the procedure used for OMC synthesis using Al-
SBA-15. The resulting OMC samples synthesized with FA was equivalent to 130% and 170% of template total pore
M AN U
volume are referred to as 1.3FA-OMC and 1.7FA-OMC, respectively.
2.2. Characterization
Powder X-ray diffraction (XRD) patterns were recorded on a Rigaku Multiplex diffractometer using a monochromatized X-ray beam from Cu Kα radiation (40 kV, 30 mA). The XRD scanning was performed under ambient conditions at step sizes of 0.01 with an accumulation time of 1.0 s. Nitrogen adsorptiondesorption isotherms were measured at 77 K using a volumetric gas sorption analyzer (Micromeritics Tristar
TE D
II). Prior to the measurement, the samples were degassed under vacuum for 3 h at 573 K. The specific surface area of the samples was calculated using the Brunauer-Emmett-Teller (BET) method with the adsorption data at a relative pressure range of 0.05 ~ 0.2. The pore size distribution was derived using the Barrett-Joyner-Halenda (BJH) method for isotherm analysis. A total pore volume was estimated from the
EP
amount of N2 adsorbed at P/P0=0.95. Transmission electron microscopy (TEM) images were recorded on a Tecnai G2 F30 and TitanTM ETEM G2 at an operating voltage of 300 kV. The X-ray photoelectron spectroscopy (XPS) experiments were performed with a Sigma Probe (Thermo VG Scientific) equipped with
AC C
a microfocused monochromator X-ray source. Temperature-programmed desorption (TPD) of NH3 was measured in a He flow using BELCAT-M
(BEL Japan) equipped with a thermal conductivity detector (TCD). Prior to the measurement, 50 mg of each
sample was placed in a quartz reactor and degassed for 3 h at 523 K. After cooling to 373 K, the NH3 gas was flowed into the reactor for 1 h at a flow rate of 30 mL min-1. Then, the free and weakly adsorbed NH3
was removed by flowing He for 1 h at 373 K. Next, the TPD profiles were measured using TCD with a continuous temperature increase (10 K min-1) under a He flow (30 mL min-1). Thermogravimetric analysis
5
ACCEPTED MANUSCRIPT
(TGA) was performed with TGA Q50 in a platinum crucible, under nitrogen atmosphere. The temperature was increased to 623 K at a ramp rate of 10 K min-1 and subsequently maintained for 60 min. Nuclear magnetic resonance (NMR) spectra were acquired in the solid state with magic angle spinning
RI PT
(MAS) using a Bruker Avance III HD 400WB spectrometer at room temperature. 31P MAS NMR spectra were recorded in a single-pulse sequence with a pulse width of 2 µs, a relaxation time of 5 s, and a spinning frequency of 12 kHz. An 85% H3PO4 aqueous solution was used as an external reference for the 31P NMR
chemical shift. 13C cross-polarization (CP) MAS NMR spectra were recorded with a contact time of 2 ms, a relaxation time of 5 s, and a spinning frequency of 10 kHz. The carbonyl carbon of glycine was used as an
M AN U
3. Results and discussion
SC
external chemical shift reference for the 13 C NMR.
3.1. OMC synthesis using the Al-SBA-15 template with excess FA
The OMC sample was synthesized using an Al-SBA-15 template by impregnating an excess amount of FA corresponding to 130% of the total pore volume (Fig. S1). As shown in Fig. 1A, the XRD pattern of the OMC product exhibits three reflection peaks at a 2θ of 0.95, 1.6, and 1.8o. The positions of the peaks were nearly same as those of the (10), (11) and (20) reflection peaks for the 2D hexagonal structure in the Al-SBA-15 template. This result
TE D
indicates that the ordered mesostructure of the carbon sample was well replicated from the silica template. However, the relative intensities between the (10) and (11) peaks for the OMC product were significantly different from those of the silica template. In particular, the intensity of the (10) peak was relatively small. This result indicates that the OMC might be composed of a mixture of CMK-3 and CMK-5 structures. It is important to note that both CMK-3 and
EP
CMK-5 could be synthesized using the Al-SBA-15 template. However, CMK-3 and CMK-5 have distinctly different XRD patterns. CMK-3, which is composed of nanorods with 2D hexagonal order, exhibits an intense (10) peak [31-
AC C
33]. On the other hand, CMK-5, which consists of a hollow nanopipe framework, exhibits negligible intensity for the (10) reflection peak due to ‘accidental extinction’ [34]. A TEM investigation confirmed that the OMC sample was composed of a mixture of nanorod-type CMK-3 and nanopipe-type CMK-5 structure (Figs. 1B and 1C). These structures were arranged with long-range hexagonal order over the entire region of the sample. However, as shown in Fig. 1D, the OMC product also has non-templated carbon species outside of the ordered mesoporous carbon
frameworks. When an excess amount of FA was impregnated into the Al-SBA-15 template, a certain amount of FA remained outside of the template. The remaining FA molecules polymerized rapidly due to the exothermic polymerization catalyzed by the strong acid sites in Al-SBA-15. The polymerized FA outside the template carbonized into non-
6
ACCEPTED MANUSCRIPT
templated carbon species. In addition to the extra FA, FA molecules, which were supposed to fill the template mesopores, could also be polymerized outside before they homogenously infiltrated into the mesopores due to the heat generated from the exothermic polymerization. A mixture of nanorods and nanopipes was obtained as the OMC
RI PT
product. From this result, it can be inferred that the selective carbon replication without external carbon species is not possible with Al-SBA-15 template via one-step impregnation. To improve the problem, the moderately weak acidity
of the template is required to provide sufficient time for the infiltration of FA into the template mesopores before the
purpose, we introduce a mineral acid impregnation method.
3.2. OMC synthesis using the PA-SBA-15 template with excess FA
SC
polymerization, and to prevent the generation of non-templated carbons during the carbonization process. For this
M AN U
We attempted to use PA-SBA-15 as an alternative method for synthesizing OMC materials. The PA-SBA-15 sample was prepared by phosphoric acid impregnation. The results of the phosphoric acid impregnation into SBA-15 are summarized in Fig. 2 and Fig. S1 in the Supplementary data. As shown in Fig. 2A, the XRD pattern of PA-SBA15 reveals that the ordered hexagonal (p6mm) mesostructure of SBA-15 was fully retained after the phosphoric acid functionalization. The N2 adsorption-desorption isotherm of PA-SBA-15 exhibits a similar appearance to that of pristine SBA-15. However, a stiff increase in the range of P/P0 between 0.6 and 0.8 was slightly shifted to lower P/P0
TE D
compared to the SBA-15 sample. The pore size distribution was determined from the adsorption branch of the isotherm by employing the BJH method. The mesopore size distribution of PA-SBA-15 is centered at 8.8 nm, which is slightly smaller than that of SBA-15 (9.1 nm) (Fig. S1B). The specific BET surface area and total pore volume of PA-SBA-15 (570 m2 g-1 and 1.00 cm3 g-1, respectively) are comparable to those of SBA-15 (690 m2 g-1 and 1.14 cm3
EP
g-1, respectively). The N2 adsorption isotherm results indicate that phosphoric acid was most likely supported on the mesopore walls of SBA-15 without pore blocking. The chemical state of the phosphate groups on the mesopore walls of PA-SBA-15 was characterized using solid
AC C
state 31P NMR spectroscopy. The 31P NMR spectrum of PA-SBA-15 contains three resonance peaks at 0.2, -11, and 23 ppm as shown in Fig. 2B. Among these peaks, the peak at 0.2 ppm was assigned to the phosphoric acids that were physically adsorbed on the mesopore walls via hydrogen bonds [35]. The peaks at -11 and -23 ppm correspond to the phosphate ester and phosphate diester, respectively, which are covalently bound to the mesopore walls [35, 36]. The
deconvolution and peak quantification of the NMR spectrum were performed to measure the amount of each species in PA-SBA-15. The molar ratio of the hydrogen-bonded phosphoric acid, phosphate ester and phosphate diester was
1.0: 10: 3.1. This result indicates that most of the phosphoric acid species were covalently bound to the silanol groups in the SBA-15 framework and formed phosphate ester groups. These species have weak Brönsted acidity [37]. The
7
ACCEPTED MANUSCRIPT
acidity of PA-SBA-15 was measured by the desorption peak in the NH3 TPD profile below 600 K (Fig. S2). This desorption peak was attributed to NH3 desorption from the Brönsted and Lewis acid sites in the framework. The peak position of PA-SBA-15 is much lower than that of Al-SBA-15. This result indicates that the strength of the acid sites
RI PT
in PA-SBA-15 are lower than those in Al-SBA-15. Using PA-SBA-15 as a template, 1.3FA-OMC was synthesized with the same amount of FA used for the AlSBA-15-templated OMC sample. The small-angle XRD pattern of the 1.3FA-OMC sample exhibits three Bragg
reflection peaks that can be indexed to the (10), (11), and (20) reflections of the 2D hexagonal structure (Fig. 3A). In
SC
contrast to the XRD pattern of Al-SBA-15-templated OMC, the intensity of the (10) reflection peak was significantly large. Therefore, the relative intensity ratio between (10) and (11) is similar to that of typical CMK-3 [32, 33]. This result indicates that 1.3FA-OMC would have a rod-type CMK-3 structure. The N2 adsorption-desorption isotherm of
M AN U
the 1.3FA-OMC sample exhibits a stiff increase in the range of P/P0 between 0.4 and 0.6, which indicates the presence of mesopores with a very uniform pore diameter (Fig. 3B). 1.3FA-OMC exhibits a narrow unimodal pore size distribution at 3.6 nm (Fig. 3B). The BET surface area and total pore volume of 1.3FA-OMC are very large (990 m2 g-1 and 0.90 cm3 g-1, respectively). The representative TEM image confirms that 1.3FA-OMC has a highly ordered hexagonal mesostructure, which is the exact inverse of the PA-SBA-15 template (Figs. 3C and D). Low magnification TEM images of 1.3FA-OMC confirmed that it was composed of nanorods without external carbon
TE D
species (Fig. S3).
It was also possible to synthesize OMC even when the amount of FA impregnated corresponded to 170% of the total pore volume of PA-SBA-15. The XRD and TEM investigation results assured that the 1.7FA-OMC product exhibited a well-ordered CMK-3 structure without non-templated carbons (Fig. S4). The mesopore diameter is very
EP
uniform with the size of 3.6 nm, which is the same as that of 1.3FA-OMC (Fig. S4C). The total pore volume and BET surface area of 1.7FA-OMC (940 m2 g-1, 0.78 cm3 g-1, respectively) were also comparable to those of CMK-3. It is important to note that faithful carbon replication was realized even when an excessive amount of FA was
AC C
impregnated within a single step. In comparison to Al-SBA-15, the acidity of PA-SBA-15 was considerably weak. The weak acidity restricted the rapid polymerization of FA molecules, which allowed the homogeneous infiltration of FA into the template mesopores. Meanwhile, the extra FA was only polymerized to a low degree, and this material was eventually vaporized during the high-temperature step, which prevented the generation of external carbon (This
will be discussed below). Therefore, a faithful CMK-3 carbon replica was obtained without external carbon species. Interestingly, the current method can result in the incorporation of P atoms in the framework of OMC. The solid
state 31P NMR spectrum of 1.3FA-OMC exhibits two resonance peaks at -6.2 and 8.5 ppm as shown in Fig. S5A. These peaks can be assigned to the PPh3 and RPh2 groups, respectively, which are most likely generated in the carbon
8
ACCEPTED MANUSCRIPT
framework. During the carbonization process, phosphate esters on the mesopore walls of the silica template are covalently bound to the carbon framework. Then, only the silica species were removed by HF while the P atoms remained in the carbon frameworks as PPh3 and RPPh2 species. The presence of P atoms in the carbon framework
RI PT
was also confirmed by XPS analysis. The XPS spectrum of 1.3FA-OMC contained two peaks centered at binding energies of 132.9 and 138 eV, which correspond to P-C and P-O bonds, respectively (Fig. S5B). From the XPS result, the content of P atoms in 1.3FA-OMC was determined to be 2.0 wt%. Such incorporation of heteroatoms into the
redox catalytic reactions [16, 38, 39].
3.3. NMR characterization of the carbonization process
SC
carbon framework can alter the intrinsic chemical properties and therefore can result in catalytic function in various
M AN U
The polymerization and carbonization process of 1.3FA-OMC were monitored by solid state 13C CP-MAS NMR. Fig. 4A shows the NMR spectra of the carbon-silica composites that were collected after thermal treatment at various temperatures. In addition, the liquid NMR spectrum of the FA monomer is shown in Fig. 4A. The NMR spectrum of the FA monomer contains five distinct peaks. Among these peaks, the peak at 56 ppm was assigned to the hydroxymethyl carbon atom (designated ‘a’ in the molecular structure of FA, shown in Fig. 4B) of FA [40]. Other four peaks in the range of 100~160 ppm correspond to the carbon atoms in the furan ring of FA.
TE D
All of the peaks from FA were broadened when the FA-containing silica was thermally treated at 333 K. The peak broadening indicates that the FA molecules were adsorbed in mesopores and started to convert into polymeric species. As the temperature increased to 373 K, the peak for the ‘a’ carbon of FA nearly disappeared, and the other peaks corresponding to the furan ring of FA became much broader. In addition, several new peaks in the range of
EP
10~40 ppm were observed. Fig. 4B shows that such a dramatic change in the NMR spectrum was due to the polymerization of FA molecules in the mesoporous silica template. During the polymerization step, the peak corresponding to the ‘a’ carbon disappeared as the alcohol groups were condensed to form C-C bonds between the
AC C
FA molecules. Instead, terminated carbons or methylene bridges between the furan rings were generated, which were confirmed by the peaks in the range of 10 ~ 40 ppm [40]. It should be noted that the rate of the FA polymerization in PA-SBA-15 is clearly different from that in Al-SBA-15 (Fig. S6). The NMR spectrum of FA-containing Al-SBA-15
obtained at 333 K is nearly same as that of PA-SBA-15 at 373 K. This result indicates that the FA molecules in the
Al-SBA-15 sample were very rapidly polymerized even at 333 K. In contrast to Al-SBA-15, the weak acid sites in PA-SBA-15 allowed the slow polymerization of FA and thus the homogenous infiltration of FA into the template. When the composite sample was thermally treated at 623 K, a broad peak at 124 ppm appeared in the NMR spectrum. This peak indicates that the polymeric FA species were transformed into aromatic carbon frameworks. In
9
ACCEPTED MANUSCRIPT
addition, there are several sharp peaks in the range of 0 ~ 50 ppm that could be assigned to alkane hydrocarbons. These hydrocarbons were likely generated from the thermal decomposition of low-degree polymerized FA outside the template. In the NMR spectrum of the composite sample collected at 1173 K, all of the peaks corresponding to
RI PT
the carbon atoms in the polymeric FAs and alkane hydrocarbons disappeared. Only a broad peak at 122 ppm corresponding to the aromatic carbon framework was observed. Based on this NMR analysis result, it can be inferred
that the FA in the PA-SBA-15 template polymerized slowly at an early stage. Therefore, there was sufficient time for homogeneous infiltration into the template mesopores. While the infiltrated FA in PA-SBA-15 were gradually
SC
polymerized and carbonized, the less polymerized extra FA outside the template decomposed and vaporized. To
confirm this assumption, we performed TGA experiments of both polymerized FA containing PA-SBA-15 and AlSBA-15 samples (Fig. S7). The FA was polymerized at 373 K. Both of the samples showed the weight loss between
M AN U
373 K and 623 K, which indicates the decomposition of polymerized FA. However, the weight loss of PA-SBA-15 composite was much larger than that of Al-SBA-15 composite. This means that the degree of the polymerization in PA-SBA-15 was lower than that in Al-SBA-15. Considering that the carbon replica obtained from PA-SBA-15 was composed of rigid carbon nanorods, the less polymerized FA species were located outside of the pores. Therefore, a faithful carbon replica was obtained using the PA-SBA-15 template regardless of the amount of FA infiltrated into
TE D
the template.
3.4. OMC Synthesis using different silica templates and mineral acids The OMC synthesis method using a phosphoric acid impregnated silica template was applied to the synthesis of OMC with various mesoporous silicas. Two different mesoporous silicas SBA-16 (cubic Im3m) and KIT-6 (cubic
EP
Ia3d) were used as templates for carbon replication. The phosphate functionalization of the silica template and carbon replication process was performed following the same procedure used for PA-SBA-15. As shown in Fig. 5A, the resulting carbon replica of phosphoric acid impregnated KIT-6 exhibits well-resolved peaks in the XRD pattern.
AC C
These peaks can be indexed to the (211), (220), and (420) reflections of the cubic Ia3d structure and the peak position is nearly same as that of the KIT-6 template [41]. The representative TEM image of the carbon replica shows a highly ordered structure (Fig. 5B). In addition, the shape of the N2 adsorption-desorption isotherm shows a typical type IV isotherm, which indicates pore condensation with hysteresis (Fig. S8). The mesopore diameter of the carbon replica was 3.8 nm, and the BET surface area and total pore volume were calculated to be 910 m2 g-1 and 1.0 cm3 g-1,
respectively. The XRD pattern for the carbon replica of phosphoric acid impregnated SBA-16 is characteristic of the cubic Im3m space group (Fig. 5C) and similar to the XRD pattern reported by Kim et al [6]. The TEM image of the carbon replica confirms that the carbon replica has a structural ordering with cubic symmetry (Fig. 5D). The BET
10
ACCEPTED MANUSCRIPT
surface area and total pore volume (780 m2 g-1 and 0.7 cm3 g-1, respectively) of the carbon replica are also comparable to that of CMK-6 [6]. Based on the results obtained from the phosphoric acid impregnation method, it was possible to extend the
RI PT
current method to different mineral acids, such as sulfuric acid. Sulfuric-acid modified SBA-15 (SA-SBA-15) was prepared by the same procedure used in PA-SBA-15, except that sulfuric acid was used instead of phosphoric acid.
The acidity of SA-SBA-15 was measured by NH3 TPD (Fig. S9). The peak of NH3-desorption from the acid sites was only observed in the low temperature region (< 500 K), which indicates that there are no strong acid sites. This result
SC
indicates that the sulfuric acid might be functionalized on the surface of SA-SBA-15 in the form of sulfate esters
which exhibit weak acidity. The small-angle XRD pattern of the OMC product synthesized with the SA-SBA-15 template exhibits three well-resolved Bragg reflection peaks that are similar to those of 1.3FA-OMC (Fig. 6A). The
M AN U
representative TEM image of the OMC product also exhibits a highly ordered hexagonal mesostructure (Fig. 6B). While the pore size distribution curve shows a sharp peak centered at 3.6 nm (Fig. S10), the BET surface area and total pore volume is 950 m2 g-1 and 0.86 cm3 g-1, respectively. In contrast to the phosphoric acid impregnation method, the elementary analysis result revealed that there are no heteroatoms in the framework. In comparison to the phosphorus, the sulfur could be easily decomposed and vaporized during the high-temperature carbonization process.
4. Conclusion
TE D
Therefore, the OMC with a pure carbon framework was obtained when SA-SBA-15 was used as a template.
We demonstrated a simple method for CMK-3 synthesis using phosphoric acid-impregnated SBA-15 as a template. The 31P NMR results indicated that impregnated phosphoric acids were functionalized by forming
EP
phosphate ester groups on the mesopore walls of SBA-15. The mild acidity of the phosphate ester in the SBA-15 template allowed the slow polymerization of FA, which prevented the generation of external carbon species even
AC C
when a large amount of FA was infiltrated. Therefore, the phosphoric acid-impregnated SBA-15 led to selective carbonization inside the mesopores and yielded highly faithful carbon replicas. The sulfuric acid could also be applied to the acid-functionalization of SBA-15 for the CMK-3 synthesis. Furthermore, this method using mineral acid impregnation could be extended to various silica templates such as KIT-6 and SBA-16. We expect that the mineral acid impregnation method would be useful for the carbon replication of various kinds of silica templates such as nanotubes, nanospheres, pure siliceous zeolites, and hierarchically porous materials.
11
ACCEPTED MANUSCRIPT
Acknowledgments This work was supported by IBS-R004-D1.
RI PT
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/xxx
[1] R. Ryoo, S. H. Joo, S. Jun, J. Phys. Chem. B 103 (1999) 7435.
SC
References
[2] S. Jun, S. H. Joo, R. Ryoo, M. Kruk, M. Jaroniec, Z. Liu, T. Ohsuna, O. Terasaki, J. Am. Chem. Soc. 122
M AN U
(2000) 10712.
[3] S. H. Joo, S. J. Choi, I. Oh, J. Kwak, Z. Liu, O. Terasaki, R. Ryoo, Nature 412 (2001) 169. [4] R. Ryoo, S. H. Joo, M. Kruck, M. Jaroniec, Adv. Mater. 13 (2001) 677. [5] J.-S. Lee, S. H. Joo, R. Ryoo, J. Am. Chem. Soc. 124 (2002) 1156.
[6] T.-W. Kim, R. Ryoo, K. P. Gierszal, M. Jaroniec, L. A. Solovyov, Y. Sakamoto, O. Terasaki, J. Mater. Chem. 15 (2005) 1560.
[7] C. Liang, K. Hong, G. A. Guiochon, J. W. Mays, S. Dai, Angew. Chem. Int. Ed. 43 (2004) 5785.
TE D
[8] F. Zhang, Y. Meng, D. Gu, Y. Yan, C. Yu, B. Tu, D. Zhao J. Am. Chem. Soc. 127 (2005) 13508. [9] M. Choi, R. Ryoo, J. Mater. Chem. 17 (2007) 4204.
[10] K.-I. Min, J.-S. Choi, Y.-M. Chung, W.-S. Ahn, R. Ryoo, P. K. Lim Appl. Cat. A: General 337 (2008) 97. [11] H. Zhou, S. Zhu, M. Hibino, I. Honma, M. Ichihara, Adv. Mater. 15 (2003) 2107.
EP
[12] N.-I. Kim, J. Y. Cheon, J. H. Kim, J. Seong, J.-Y. Park, S. H. Joo, K. Kwon, Carbon 72 (2014) 354. [13] A. Vinu, P. Srinivasu, M. Takahashi, V. V. Balasubramanian, K. Ariga, Micropor. Mesopor. Mater. 153
AC C
(2001) 44.
[14] H. I. Lee, J. H. Kim, D. J. You, J. E. Lee, J. M. Kim, W.-S. Ahn, C. Pak, S. H. Joo, H. Chang, D. Seung, Adv. Mater. 20 (2008), 757.
[15] J. Kim, M. Choi, R. Ryoo, Bull. Korean. Chem. Soc. 29 (2008) 413. [16] D.-S. Yang, D. Bhattacharjya, S. Inamdar, J. Park, J.-S. Yu, J. Am. Chem. Soc. 134 (2012) 16127. [17] X. Zhao, Q. Zhang, B. Zhang, C.-M. Chen, A. Wang, T. Zhang, D. S. Su, J. Mater. Chem. 22 (2012) 4963.
[18] D. Zhang, Y. Hao, L. Zheng, Y. Ma, H. Feng, H. Luo, J. Mater. Chem. A 1 (2013) 7584. [19] T.-W. Kim, I.-S. Park, R. Ryoo, Angew. Chem. Int. Ed. 42 (2003) 4375.
12
ACCEPTED MANUSCRIPT
[20] J. Lee, S. Yoon, T. Hyeon, S. M. Oh, K. B. Kim, Chem. Commun. (1999) 2177. [21] O. J. A. Schueller, S. T. Brittain, G. M. Whitesides, Adv. Mater. 9 (1997) 477. [22] G. Tondi, A. Pizzi, R. Olives Maderas. Ciencia y tecnologia 10 (2008) 219.
RI PT
[23] T.-W. Kim, L. A. Solovyov, J. Mater. Chem. 16 (2006) 1445. [24] K. P. Gierszal, M. Jaroniec, T.-W. Kim, J. Kim, R. Ryoo, New J. Chem. 32 (2008) 981. [25] Z. Tian, M. A. Snyder, Langmuir 30 (2014) 12411. [26] Z. Tian, M. A. Snyder, Langmuir 30 (2014) 9828.
SC
[27] A.-H. Lu, W.-C. Li, W. Schmidt, W. Kiefer, F. Schüth, Carbon 42 (2004) 2939.
[28] P. A. Bazula, A.-H. Lu, J.-J. Nitz, F. Schüth, Micropor. Mesopor. Mater. 108 (2008) 266.
[29] A. Galarneau, H. Cambon, F. D. Renzo, R. Ryoo, M. Choi, F. Fajula, New J. Chem. 27 (2003) 73.
M AN U
[30] T.-W. Kim, F. Kleitz, B. Paul, R. Ryoo, J. Am. Chem. Soc. 127 (2005) 7601. [31] I.-S. Park, M. Choi, T.-W. Kim, R. Ryoo, J. Mater. Chem. 16 (2006), 3409. [32] A. B. Fuertes, Micropor. Mesopor. Mater. 67 (2004) 273.
[33] J. Roggenbuck, G. Koch, M. Tiemann, Chem. Mater. 18 (2006) 4151.
[34] K. Lund, N. Muroyama, O. Terasaki, Micropor. Mesopor. Mater. 128 (2010) 71. [35] L.-Q. Wang, W. D. Samuels, G. J. Exarhos, B. I. Lee, Z. Cao, J. Mater. Chem. 8 (1998) 165.
TE D
[36] G. D. Cody, B. Mysen, G. Sághi-Szabó, J. A. Tossell, Geo. et Cos. Acta 65 (2001) 2395. [37] T. L. Lamoureaux, D. H. Lee, Chem. Commun. 47 (2011) 8623. [38] J. Wu, Z. Yang, X. Li, Q. Sun, C. Jin, P. Strasser, R. Yang, J. Mater. Chem. A 1 (2013) 9889. [39] P. Karthika, N. Rajalakshmi, K. S. Dhathathreyan, J. Nanosci. Nanotech. 13 (2013) 1746.
EP
[40] L.-T. Cheng, W. J. Tseng, J. Polym. Res. 17 (2010) 391.
AC C
[41] F. Kleitz, S. H. Choi, R. Ryoo, Chem. Commun. (2003) 2136.
13
ACCEPTED MANUSCRIPT
M AN U
SC
RI PT
Figures
Fig. 1. (A) XRD pattern and (B-D) TEM images of the OMC product obtained using the Al-SBA-15 template. (B) and (C) are nanorods and nanopipes, respectively, in the OMC product. White arrows in (D) indicate external carbon
AC C
EP
TE D
species. For comparison, the XRD pattern of Al-SBA-15 is included in (A).
Fig. 2. (A) Powder XRD pattern and (B) Solid state 31P MAS NMR spectrum of PA-SBA-15. For comparison, the XRD pattern of SBA-15 is included in (A).
14
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Fig. 3. (A) Powder XRD pattern, (B) N2 adsorption-desorption isotherm and mesopore size distribution corresponding to the adsorption branch, and (C) TEM image of 1.3-FA-OMC. For comparison, a TEM image of PA-
AC C
EP
SBA-15 is shown in (D).
15
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Fig. 4. (A) Liquid state 13C NMR spectrum of the FA monomer and solid state 13C CP-MAS NMR spectra of the
AC C
EP
TE D
carbon-silica composite samples collected at various temperatures and (B) scheme for FA polymerization.
16
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Fig. 5. Powder XRD patterns and TEM images of carbon replica of (A, B) KIT-6 and (C, D) SBA-16. For
AC C
EP
comparison, the XRD pattern of the KIT-6 and SBA-16 template was included.
Fig. 6. (A) Powder XRD pattern and (B) TEM image of OMC obtained from Sulfuric-acid modified-SBA15.
17
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
18
ACCEPTED MANUSCRIPT
Highlights
AC C
EP
TE D
M AN U
SC
RI PT
Mineral acid could be simply anchored on silica templates through ester bonding. > The functionalized silicas showed moderate acidity for faithful carbon replication. > The moderate acidity led to selective carbon formation inside the template pores. > External carbon deposition was prevented despite using carbon precursor excessively. > The present method could be extended for various mesoporous silica templates.
ACCEPTED MANUSCRIPT
Synthesis of Mesoporous Carbons using Silica Templates
RI PT
Impregnated with Mineral Acids
Yongbeom Seo,† Kyoungsoo Kim,‡ Younjae Jung,† and Ryong Ryoo‡,†,*
Department of Chemistry, Korea Advanced Institute of Science and Technology (KAIST),
SC
†
Daejeon 305-701, Korea
Center for Nanomaterials and Chemical Reactions, Institute for Basic Science (IBS), Daejeon
M AN U
‡
305-701, Korea
*Corresponding author: Ryong Ryoo (E-mail address: [email protected]; Tel.: +82 42 350
AC C
EP
TE D
2830)
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
TE D
Figure S1. (A) N2 isotherm and (B) pore size distribution of SBA-15 and PA-SBA-15. Isotherms and pore size distributions of samples are vertically offset by 400 cm3 g-1 and 5.0 cm3 g-1, respectively. The BET surface area and total pore volume of Al-SBA-15 are
AC C
EP
625 m2 g-1 and 1.13 cm3 g-1, respectively.
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Figure S2. NH3 TPD profiles of PA-SBA-15 and Al-SBA-15. The result for PA-SBA15 shows an additional peaks above 800 K, but it is caused by thermal decomposition of
EP
phosphate esters and condensation of silanol groups. For comparison, TPD profile of
AC C
PA-SBA-15 without NH3 is included.
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
Figure S3. Low-magnification TEM images of 1.3FA-OMC.
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
EP
Figure S4. (A) Powder XRD pattern, (B) TEM image, and (C) N2 adsorptiondesorption isotherm and mesopore size distribution corresponding to the adsorption
AC C
branch of 1.7FA-OMC.
31
M AN U
Figure S5. (A) Solid state
SC
RI PT
ACCEPTED MANUSCRIPT
P MAS NMR spectrum and (B) XPS spectrum of 1.3FA-
OMC. ‘*’ in (A) indicates spinning side bands. The red and blue curves in (B) indicate
AC C
EP
TE D
results of spectral analysis by Gaussian deconvolution.
Figure S6. Solid state
13
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
C CP-MAS NMR spectrum of carbon/Al-SBA-15 composite
AC C
EP
TE D
sample collected at 333 K.
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
Figure S7. TGA curves for the polymerized FA-containing silica templates. The temperature was increased to 623 K in N2 flow at a ramp rate of 10 K min-1 and subsequently maintained for 60 min.
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Figure S8. N2 isotherm and pore size distribution of SBA-16 and KIT-6 template and
AC C
their carbon replicas. Isotherms and pore size distributions of samples are vertically offset by 100 cm3 g-1 and 1.5 cm3 g-1, respectively.
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Figure S9. NH3 TPD profile of SA-SBA-15. The result for SA-SBA-15 shows an
TE D
additional peaks above 800 K, but it is caused by condensation of silanol groups and
AC C
EP
thermal decomposition of sulfate ester groups.
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Figure S10. N2 isotherm and pore size distribution of carbon replica of SA-SBA-15.
AC C
EP
TE D
.