Microporous and Mesoporous Materials 197 (2014) 77–82
Contents lists available at ScienceDirect
Microporous and Mesoporous Materials journal homepage: www.elsevier.com/locate/micromeso
Effect of silica sources on nanostructures of resorcinol–formaldehyde/ silica and carbon/silicon carbide composite aerogels Yong Kong a,b, Ya Zhong a, Xiaodong Shen a,⇑, Sheng Cui a, Maohong Fan b,⇑ a b
College of Materials Science and Engineering, Nanjing Tech University, Nanjing 210009, PR China Department of Chemical and Petroleum Engineering, University of Wyoming, Laramie, WY 82071, USA
a r t i c l e
i n f o
Article history: Received 7 March 2013 Received in revised form 9 February 2014 Accepted 23 May 2014 Available online 5 June 2014 Keywords: Aerogel Pore structure Microstructure Supercritical fluid drying Thermal insulation
a b s t r a c t The subject of this paper is the investigation of the effect of silica sources on microstructure of resorcinol– formaldehyde/silica composite (RF/SiO2) and carbon/silicon carbide composite (C/SiC) aerogels. Hybrid silica sources (HSS) were composed of 3-(aminopropyl)triethoxysilane (APTES) and tetraethoxysilane (TEOS) with different molar ratio. RF/SiO2 aerogel was obtained by a single-step sol–gel process followed by supercritical fluid drying (SCFD). C/SiC aerogel was formed from RF/SiO2 aerogel after carbothermal reduction. Scanning electron microscopy (SEM) and N2 adsorption/desorption were used to investigate the evolution of morphology and pore structures of aerogels. X-ray diffraction (XRD) and transmission electron microscopy (TEM) demonstrated that the as-prepared C/SiC aerogel was composed of carbon nanoparticle and a-SiC nanocrystal. The microstructure was hugely affected by the component of HSS. When the molar fraction of APTES in HSS was 60%, RF/SiO2 and C/SiC aerogels possessed the highest surface area and pore volume and the lowest thermal conductivity. Ó 2014 Elsevier Inc. All rights reserved.
1. Introduction Aerogels are nanoporous materials with high surface areas, high porosities, low densities and small pore size [1–6]. Among all kinds of aerogels that have been explored to this day, carbon/silicon carbide composite (C/SiC) aerogels have excellent properties, including chemical and thermal stability, high conductivity, low thermal conductivity, high surface area and high porosity [7,8]. Hence, they can be used as adsorbents, thermal insulators, supports of catalysts or electrode materials [9–14]. In the past, binary carbonaceous silica aerogels which were commonly prepared using tetraethoxysilane (TEOS) or tetramethylorthosilicate (TMOS) as the silica sources were used as precursors of porous C/SiC by carbothermal reduction [6,8,15–19]. However, the sol–gel processes employed to form hybrid gels were complicated or timeconsuming, and too many catalysts were required. Moreover, the silica sol and carbonaceous sol had to be prepared separately to form carbon–silica hybrid gels. Therefore, we purposed a facile synthesis of carbonaceous silica aerogels by a single-step sol–gel reaction followed by supercritical drying [7]. In this method, 3-(aminopropyl)triethoxysilane (APTES) was used as the silica source, resorcinol–formaldehyde/silica composite (RF/SiO2) gel ⇑ Corresponding authors. Tel.: +86 25 83587235; fax: +86 25 83221690 (X. Shen). Tel.: +1 307 766 5633; fax: +1 307 766 6777 (M. Fan). E-mail addresses:
[email protected] (X. Shen),
[email protected] (M. Fan). http://dx.doi.org/10.1016/j.micromeso.2014.05.032 1387-1811/Ó 2014 Elsevier Inc. All rights reserved.
was synthesized in one pot by simply mixing the monomers. The as-prepared RF/SiO2 and C/SiC aerogels possessed high surface area and pore volume. We have also developed another technique using TEOS as silica source to synthesize RF/SiO2 aerogels [20]. Recently, it was found that different silica sources (TEOS and APTES) leaded to very different microstructures. Introducing APTES into the R-FTEOS system would strongly affect the sol–gel behavior. Therefore, the effect of silica sources which were composed of APTES and TEOS with different molar ratio on nanostructures of RF/SiO2 aerogels and C/SiC aerogels was studied in this work. Also, thermal conductivity of the resulting C/SiC aerogels was investigated.
2. Experimental APTES, TEOS, resorcinol (R), formaldehyde (F, 37% w/w aqueous solution) and anhydrous alcohol (EtOH) were used as raw materials. All chemicals were analytical grade and used as received without further purification. APTES and TEOS were used as the hybrid silica source. The molar fractions of APTES in the hybrid silica source (HSS) are 20%, 40%, 60% and 80%, and the corresponding samples are denoted as RFA20, RFA40, RFA60 and RFA80, respectively. Samples prepared only using APTES (100% APTES content in HSS) and TEOS (0% APTES content in HSS) as silica source which were denoted as RFA and RFT were also involved to comparison. The synthesis route of RF/SiO2 and C/SiC aerogels is presented in
78
Y. Kong et al. / Microporous and Mesoporous Materials 197 (2014) 77–82
Scheme 2. The surface structure of the framework with APTES involved.
Scheme 1. Synthesis of RF/SiO2 and C/SiC aerogels.
Fig. 3. Mass loss and linear shrinkage of C/SiC relative to RF/SiO2.
Fig. 1. Influence of the contents of APTES in the silica source on the gelation time.
Fig. 4. XRD data of C/SiC aerogel of RFA and RFA20.
Fig. 2. Effect of silica source on the density of RF/SiO2 and C/SiC aerogels.
Scheme 1. R, F, SS, W and EtOH were mixed in a pot at room temperature, with R:F:HSS:EtOH:W prepared at a molar ratio of 1:2:1:60:2. Then the compounds were poured into polypropylene moulds and the gels were formed at 60 °C. The gelation time was
recorded when the solution did not move with 45 degree slope. After gelation the wet gels were aged at 75 °C for 24 h and simultaneously washed with ethanol every 8 h to remove water and residual chemicals. After solvent exchange, the alcohol gels were dried using supercritical CO2 fluid drying (SCFD, HELIX 1.1 system with 2 L vessel, Applied Separations, Inc., Allentown, PA, USA) to form RF/SiO2 aerogels. The procedure of SCFD is as follows: the alcohol gels were first put into the vessel, the pressure of the vessel was raised up to 10 MPa by pumping the liquid CO2 into the vessel
Y. Kong et al. / Microporous and Mesoporous Materials 197 (2014) 77–82
Fig. 5. TEM image (a) and HRTEM image (b) of C/SiC aerogels of RFA20.
79
with a CO2 pump. Ethanol was first replaced by liquid CO2 with a flow rate of 20 L/min under room temperature (20 °C) till there was no ethanol out from the outlet, and then the temperature was raised to 50 °C for supercritical extraction with a flow rate of 10 L/min and maintained at that level for 6 h to ensure that the ethanol was displaced by CO2 thoroughly. In these processes, the pressure of the vessel was maintained at 10 MPa all the time. At the end, the vessel was depressurized to atmosphere pressure at a flow rate of 5 L/min to get RF/SiO2 aerogels. The thermal treatment was performed in a tube furnace (72 and 80 mm inner and outer diameters of tube, 120 mm heating zone). RF/SiO2 aerogels were converted to C/SiC aerogels by carbothermal reduction in flowing argon (100 ± 10 ml/min) at 1500 °C for 5 h. Apparent density (q) was calculated from the weight and the physical dimensions of the samples. The morphology of the specimens was surveyed by LEO 1530VP scanning electron microscope (SEM). The phase composition of the sample was evaluated by ARL ARLX’TRA X-ray diffraction (XRD) using a Cu-Ka radiation. Transmission electron microscopy (TEM) was conducted using a JEOL JEM-2010 electron microscope. Surface areas, pore volume and pore distribution were measured by nitrogen adsorption/ desorption porosimetry by using a Micromeritics ASAP2020 surface area and pore distribution analyzer after the samples were degassed in a vacuum at 90 °C for 6 h. The specific surface area (r) was calculated using Brunaur–Emmett–Teller (BET). t-Plot model was used for calculation of micropore volume. By using the Barrett–Joyner–Halenda (BJH) model, the pore-size distribution was derived from the desorption branch of isotherms, and
Fig. 6. SEM images of RF/SiO2 and C/SiC aerogels: (a) RF/SiO2 of RFA60, (b) C/SiC of RFA60, (c) RF/SiO2 of RFA, (d) C/SiC of RFA, (e) RF/SiO2 of RFT, (f) C/SiC of RFT.
80
Y. Kong et al. / Microporous and Mesoporous Materials 197 (2014) 77–82
the pore volume was estimated from the adsorbed amount at a relative pressure p/p0 of 0.99. Thermal conductivity is measured by using a Hot Disk TPS 2500 S analyzer.
Table 1 Pore structure data of RF/SiO2 and C/SiC aerogels.
3. Results and discussion Fig. 1 shows the influence of silica source on the gelation time. The gelation time presents an inverse trend with the increase of the APTES content. In the system, the effect of APTES is not just that of a reactant, it also acts as a catalyst. The amine groups in APTES are an internal catalyst for the condensation of resorcinol and formaldehyde [21]. It was found in the experiments that the sol– gel reaction did not take place without APTES in the R-F-APTES and R-F-TEOS-APTES system, and a small quantity of APTES could initiate the polymerization. It is attributed to the formation of intramolecular hydrogen bonds in the hydrolyzed species of APTES [22]. Such a cyclic intermediate could provide an energetically favorable reaction path, thereby accelerating the hydrolysis and polymerization in the sol–gel process. Fig. 2 shows the effect of silica source on the apparent density. It is found that RFT shows the lowest density for RF/SiO2 aerogels. The increase from q(RF/SiO2) to q(C/SiC) is the largest for RFT. It reveals that the microstructure of RFT is affected by the thermal treatment gravely. After introducing APTES into the system, the density of RF/SiO2 and C/SiC aerogels presents a growing tendency while increasing the APTES content in the HSS. For RF/SiO2 aerogels, it results from the contribution of the aminopropyl groups. Three alkoxy groups in APTES can hydrolyze and polymerize to generate a network with aminopropyl groups on the surface of framework (Scheme 2) [23]. Therefore, there are more organic groups in the framework when more APTES is introduced into the silica source, which lead to the increase of q(RF/SiO2). For C/SiC aerogels, the competition between mass loss and shrinkage in the thermal treatment process decided the variation from q(RF/SiO2) to q(C/SiC). Fig. 3 shows the mass loss and linear
Sample
BET surface area (m2/g)*
Mesopore volume (cm3/g)*
Micropore volume (cm3/g)*
RFT RFA20 RFA40 RFA60 RFA80 RFA
527 282 358 410 386 384
2.435 1.164 1.402 1.861 1.557 1.473
0.0450 0.0221 0.0282 0.0384 0.0332 0.0325
[564] [580] [877] [946] [910] [892]
[2.587] [1.363] [2.145] [2.744] [2.668] [2.603]
[0.0477] [0.042] [0.100] [0.106] [0.102] [0.098]
*
The data out and in bracket are the parameters of RF/SiO2 and C/SiC aerogels, respectively.
shrinkage of C/SiC relative to RF/SiO2. Both mass loss and linear shrinkage increased along with the increase of APTES in HSS. It is reasonable because a portion of pore volume is occupied by the aminopropyl and other organic groups. More organic groups lead to more mass loss and more shrinkage. Although C/SiC aerogels presented significant mass loss (more than 59% for all the samples) and linear shrinking (more than 30% for all the samples) relative to RF/SiO2 aerogels, they still preserved monolithic morphology. As seen from Fig. 2, the increase of the density from RF/SiO2 to C/SiC maintains at a low level when the molar fraction of APTES is in the range of 60% to 80%. This leads to larger porosity, which can be seen from the pore structure data from N2 adsorption/desorption test. It suggests that introducing APTES into the system can weaken the influence of thermal treatment on the pore structure. Fig. 4 shows the XRD patterns of C/SiC aerogels of RFA and RFA20. It was demonstrated that SiC in C/SiC aerogels of RFA and RFT is a and b phase, respectively [7,20]. SiC in C/SiC aerigels of RFA20 was also a phase (PDF#29-1131) although some diffraction peaks of RFA20 were not very obvious relative to that of RFA. It could be explained by the appearance of Si-C bond that already existed in the carbonaceous silica aerogels, which could accelerate the growth of nanocrystals and lead to the formation of a-SiC.
Fig. 7. N2 adsorption/desorption isotherms of RF/SiO2 (a) and C/SiC (b) aerogels.
81
Y. Kong et al. / Microporous and Mesoporous Materials 197 (2014) 77–82 Table 2 Pore structure data of different RFA60 samples. Sample
BET surface area (m2/g)*
Mesopore volume (cm3/g)*
Micropore volume (cm3/g)*
RFA60-1 RFA60-2 RFA60-3
410 [946] 416 [955] 399 [938]
1.861 [2.744] 1.867 [2.747] 1.849 [2.736]
0.0384 [0.106] 0.0387 [0.107] 0.0382 [0.106]
*
The data out and in bracket are the parameters of RF/SiO2 and C/SiC aerogels, respectively.
Fig. 8. Pore-size distribution of RF/SiO2 (a) and C/SiC (b) aerogels.
Fig. 9. Cumulative pore volume curves of RF/SiO2 aerogels.
Fig. 5 shows the TEM image and high-resolution transmission electron image (HRTEM) of C/SiC aerogels of RFA20. TEM image reveals that SiC nanoparticles distribute in C/SiC composites equably. As observed from the HRTEM, the lattice fringe, with spacing of approximately 0.235 nm, corresponds to the 103 crystal plane of a-SiC (PDF#29-1131). In combination with the XRD patterns and
the HRTEM image reveal that the as-prepared C/SiC is composed of carbon nanoparticles and a-SiC nanocrystals. Fig. 6 presents the SEM images of RF/SiO2 and C/SiC aerogels prepared using different content of APTES. It is found the variation of microstructure for RFT between RF/SiO2 and C/SiC aerogel is more obvious than others, which also demonstrates that introducing APTES into the system weakens the influence of thermal treatment on the pore structure. For RF/SiO2 aerogels, all samples exhibit the disordered, porous structures like a typical colloidal gel [24]. The local agglomeration of RF/SiO2 aerogel for RFA is the most obvious. The nanoparticles of C/SiC aerogels of RFT present the similar structure as the RF/SiO2 aerogels. However, the nanoparticles of C/SiC aerogels are non-spherical and indistinguishable after carbothermal reduction after introducing APTES into the system. As mentioned above, it may result from the Si-C bond, which accelerates the carbothermal reduction reaction. The pore structure of the samples are evaluated by using nitrogen adsorption/desorption test. Fig. 7 shows the nitrogen adsorption/desorption isotherms of all the RF/SiO2 and C/SiC aerogels. They are all type IV curves with H1 hysteresis loop in the IUPAC classification, suggesting that they are mesoporous. The pore structure data are summarized in Table 1. RFA60 possessed the highest surface area and pore volume for both RF/SiO2 and C/SiC aerogels because overmuch surface organic group could block the pores space to some degree. The surface area and pore volume increase after carbothermal reduction, due to the removing of organic groups during thermal treatment. However, these increase for RFT after is small due to the huge increase of density. Fig. 8 shows the pore-size distribution curves of the samples. Micropores (
