Synthesis of mesoporous tin dioxide via sol-gel process assisted by resorcinol–formaldehyde gel

G Model PARTIC-1064; No. of Pages 7

ARTICLE IN PRESS Particuology xxx (2017) xxx–xxx

Contents lists available at ScienceDirect

Particuology journal homepage: www.elsevier.com/locate/partic

Synthesis of mesoporous tin dioxide via sol-gel process assisted by resorcinol–formaldehyde gel Kornkamol Banjerdteerakul, Paravee Vas-Umnuay, Varong Pavarajarn ∗ Center of Excellence in Particle Technology, Department of Chemical Engineering, Faculty of Engineering, Chulalongkorn University, Thailand

a r t i c l e

i n f o

Article history: Received 19 April 2017 Received in revised form 24 June 2017 Accepted 11 July 2017 Available online xxx Keywords: Tin dioxide Mesoporous Resorcinol–formaldehyde gel Synthesis

a b s t r a c t Tin dioxide is a useful n-type oxide semiconductor used in a variety of applications owing to its superior optical, electrical, and multifunctional properties. Here, we used a network of resorcinol–formaldehyde (RF) gel to synthesize mesoporous tin dioxide via a sol–gel process. The effects of various synthesis parameters on the morphology and mesoporosity of the obtained product were investigated, including aging time of the RF gel, tin-to-formaldehyde molar ratio, resorcinol-to-carbonate molar ratio, and the aging time of the tin/RF mixed gel. Our experimental results showed that the interaction between the network of the RF gel and tin-containing sol is a key factor that affected the structural strength of the porous network and the porosity of the final product. Through control of the interactions in the tin/RF mixed gel we obtained porous tin dioxide materials that could be effectively used to form large-surface area films with desirable mesoporous properties. © 2017 Chinese Society of Particuology and Institute of Process Engineering, Chinese Academy of Sciences. Published by Elsevier B.V. All rights reserved.

Introduction Owing to its remarkable optical, electrical, and mechanical properties, tin dioxide (SnO2 ) is considered a particularly useful n-type oxide semiconductor, with a broad energy band gap of 3.6–3.8 eV. Although bulk SnO2 is already used in many applications, mesoporous SnO2 has potential for use in fields, such as catalysis (Zhang, Huang, & Popov, 2010), gas sensing (Jin, Zhou, Jin, Savinell, & Liu, 1998; Shimizu, Jono, Hyodo, & Egashira, 2005), photoelectrochemical cells (Toupance, Babot, Jousseaume, & Vilaca, 2003), and lithium battery anodes (Wen, Wang, Zhang, & Li, 2007). Various approaches for preparing mesoporous SnO2 have been reported, including hydrothermal synthesis (Fujihara et al., 2004; Wen et al., 2007), spray pyrolysis (Hieda, Hyodo, Shimizu, & Egashira, 2008), chemical vapor deposition (Bruno, Pijolat & Lalauze, 1994), precipitation (Song & Kim, 2000), and sol–gel processes (Chen & Liu, 1999; Emons, Li & Nazar, 2002; Toupance et al., 2003). Among these techniques, sol–gel processes are particularly attractive because of their simplicity, ambient temperature processing conditions, and formation of products with good homogeneity and narrow size distributions.

∗ Corresponding author. E-mail address: [email protected] (V. Pavarajarn).

To induce mesoporosity within a material, a surfactant is often incorporated as a template (Ahmed et al., 2008; Yu & Frech, 2002; Zhou, Lu, Ke, & Li, 2003). However, common surfactants used in the syntheses of mesoporous SnO2 result in the uncontrollable formation of microporous structures and a mixture of micro- and mesoporous structures (Li et al., 2008; Xi et al., 2008). Material properties show a strong dependence on the structure, and potential applications of mesoporous SnO2 require materials with high crystallinity and high surface areas. In general, surfactants decompose at relatively low temperatures and are ineffective for preventing the collapse of pores during calcination. However, calcination at high temperatures is necessary to induce high crystallization in mesoporous tin dioxide. According to the previous studies, calcination at 400 ◦ C is insufficient to completely remove the surfactant from the product (Chen & Liu, 1999; Wang, Ma, Sun, & Li, 2001; Jain, Rashmi, & Lakshmikumar, 2005). As a result, crystallization of SnO2 is hindered owing to the presence of residual surfactant. Crystallization can be enhanced by increasing the calcination temperature to 600 ◦ C, to completely remove the surfactant (Jain et al., 2005); however, this generally results in an unstable structure and loss of mesoporosity, which reduces the surface area (Guo et al., 2011; Ahmad, Abu Bakar, & Jusliha, 2016). These factors have limited the practical applications of mesoporous SnO2 (Toupance et al., 2003; Hyodo, Nishida, Shimizu, & Egashira, 2001). Resorcinol–formaldehyde gel (RF gel) has been recognized as a template for mesoporous carbon aerogels that have surface areas

https://doi.org/10.1016/j.partic.2017.07.006 1674-2001/© 2017 Chinese Society of Particuology and Institute of Process Engineering, Chinese Academy of Sciences. Published by Elsevier B.V. All rights reserved.

Please cite this article in press as: Banjerdteerakul, K., et al. Synthesis of mesoporous tin dioxide via sol-gel process assisted by resorcinol–formaldehyde gel. Particuology (2017), https://doi.org/10.1016/j.partic.2017.07.006

G Model PARTIC-1064; No. of Pages 7

ARTICLE IN PRESS K. Banjerdteerakul et al. / Particuology xxx (2017) xxx–xxx

2

as high as 1000 m2 /g. The decomposition temperature of this gel is approximately 500 ◦ C, which makes it suitable for synthesis of high-temperature materials (Babic, Kaluderovic, Vracar, & Krstajic, 2004; Jirglova & Maldonado-Hodar, 2010; Luyjew, Tonanon, & Pavarajarn, 2008). RF gel is a product derived from the polycondensation between resorcinol and formaldehyde catalyzed by sodium carbonate (Pekala, 1989). The process is not a direct templating because the interactions between the RF mixture and precursors for sol–gel synthesis of a desired material usually occur (Chantam & Pavarajarn, 2012). For this reason, RF gel is a potentially interesting alternative template for forming stable and well-crystallized mesoporous SnO2 at high calcination temperatures. Herein, mesoporous SnO2 was synthesized with an RF gel template. The aim of this work was to investigate the interactions between the tin-precursor and the RF gel by studying the effects of synthesis conditions on the characteristics of the mesoporous tin dioxide. Our approach offers a way to systematically and precisely control the mesoporosity of SnO2 .

Fig 1. Typical FTIR spectra of SnCl4 (a), formaldehyde (b), tin-containing sol (c), neat RF gel (d), and Sn/RF composite (e).

Experimental Preparation of preformed tin-containing sol The tin-precursor, i.e., tin chloride pentahydrate (Sigma–Aldrich, Milwaukee, WI), reacted vigorously with the RF gel, resulting in rapid solidification of the mixture; hence, a tincontaining sol was first prepared by dissolving the tin-precursor in formaldehyde (CH2 O, Asia Pacific Specialty Chemical, Sydney, Australia). The molar ratio of Sn/CH2 O was varied from 0.003 to 0.06. The solution was stirred for 15 min before aging at room temperature for a predetermined period of time.

size distribution of the samples. The average pore size was obtained by the Barrett–Joyner–Halenda (BJH) method. Fourier transform infrared spectroscopy (FTIR, Nicolet 6700, Waltham, MA) was used to identify the chemical functional groups in the gels. The microstructure and morphology of the samples were examined by scanning electron microscopy (SEM, JEOL JSM6400, Tokyo, Japan). The crystalline phases of the products were investigated by X-ray diffraction (XRD, Bruker AXS D8 Advance, Karlsruhe, Germany), while the thermal decomposition behavior was determined by thermogravimetric analysis (TGA, MettlerToledo TGA/DSC1 STARe, Greifensee, Switzerland).

Preparation of resorcinol–formaldehyde gel Resorcinol (C6 H6 O2 , Asia Pacific Specialty Chemical, Sydney, Australia) was dissolved in deionized water (H2 O) and stirred until C6 H6 O2 was completely dissolved. The C6 H6 O2 /H2 O molar ratio was maintained at 0.15. Next, sodium carbonate (Na2 CO3 , Asia Pacific Specialty Chemical, Sydney, Australia), used as the catalyst for RF gel formation, was added to the solution at C6 H6 O2 /Na2 CO3 molar ratio in the range of 50–300. CH2 O was then added to the mixture in the last step to give an C6 H6 O2 /CH2 O molar ratio of 0.5. The mixture was stirred until the dissolution was complete before being aged at room temperature for 1–4 h. Synthesis of porous tin dioxide Ethanol (C2 H6 O, VWR International S.A.S., France) was added to the aged-RF gel with an C2 H6 O-to-C6 H6 O2 molar ratio of 5.0 and was stirred for 15 min. The preformed tin-containing sol was added to the RF gel. The Sn/C6 H6 O2 molar ratio was maintained at 0.08. The mixture was stirred until it became homogeneous and then aged at room temperature for 0–5 days. The aged gel was then immersed in t-butanol for 3 days, with replacement of fresh t-butanol each day. The washed gel was frozen at −40 ◦ C for 24 h before being freeze-dried for another 24 h. Porous SnO2 was obtained after calcination at a controlled temperature in the range 400–700 ◦ C for 6 h with a heating rate of 5 ◦ C/min to remove the carbon from the composite. Product characterization Nitrogen adsorption/desorption isotherms were measured using a Brunauer–Emmett–Teller analyzer (BET, Belsorp mini II, Tokyo, Japan) to investigate the specific surface area and pore-

Results and discussion Product characteristics Without any special treatments, the RF gel mixture reacted violently with the tin-precursor, causing the gel to solidify almost instantaneously. Hence, a tin-containing sol was formed prior to the addition to the RF gel to decrease the activity of the tin-precursor and enable a uniform distribution of the tin within the Sn/RF mixture. To investigate the interaction between the tin-containing sol and the RF gel template, FTIR spectra of the samples at various stages of the synthesis are compared in Fig. 1. FTIR spectrum of the tin-containing sol showed the presence of formaldehyde. The absorption band of the tin-precursor at 1633 cm−1 merged into a broad band with the signal at 1640 cm−1 arising from formaldehyde. The band from the tin-precursor at 862 cm−1 could also be seen in the tin-containing sol. When the tin-containing sol was mixed with the RF gel, the obtained Sn/RF composite shared the main features of the RF gel, i.e., the C C stretching vibration of the aromatic rings at 1614 cm−1 , the CH2 stretching vibration of the methylene bridge at 1477 cm−1 , and the C O C stretching vibrations of the methylene ether bridge at 1222 and 1092 cm−1 (Pekala, 1989; Poljansek & Krajnc, 2005). The signal at 862 cm−1 from the tin-precursor was no longer detectable in the spectrum of the Sn/RF composite. Instead, a signal at 880 cm−1 corresponding to Sn O C (Krissanasaeranee, Supaphol, & Wongkasemjit, 2010) appeared as a shoulder of the absorption band at 896 cm−1 of the RF gel. This result indicates that an interaction existed between the tin-containing sol and the RF gel. Owing to overlapping of the bands, the intensity of the band at 880 cm−1 was determined by processing the spectrum by Fourier self-deconvolution.

Please cite this article in press as: Banjerdteerakul, K., et al. Synthesis of mesoporous tin dioxide via sol-gel process assisted by resorcinol–formaldehyde gel. Particuology (2017), https://doi.org/10.1016/j.partic.2017.07.006

G Model PARTIC-1064; No. of Pages 7

ARTICLE IN PRESS K. Banjerdteerakul et al. / Particuology xxx (2017) xxx–xxx

Fig. 2. Thermogravimetric analysis results of tin-containing sol (a), neat RF gel (b), and Sn/RF composite (c). Complementary differential scanning calorimetry curves are shown in inset.

Fig. 2 shows the thermogravimetric analysis results of the tincontaining sol, neat RF gel, and Sn/RF composite. The analysis was performed in oxygen at a heating rate of 10 ◦ C/min. The inset shows the differential scanning calorimetry (DSC) curves from the same analysis. The tin-containing sol showed an endothermic peak at approximately 100 ◦ C corresponding to evaporation of the solvent. The mass loss of the neat RF gel sample occurred in two steps at 380 and 450 ◦ C, attributed to breaking of C O and C H bonds in the RF network, respectively (Lin & Ritter, 1997). However, only a single sharp exothermic peak was observed at approximately 450 ◦ C in the DSC curve. For the Sn/RF composite, the major thermal decomposition occurred at approximately 240 ◦ C, followed by another thermal event similar to that of the neat RF gel at approximately 400 ◦ C. This result suggests that the Sn/RF composite consisted of two components, i.e., a compound resulting from the interaction between the tin-containing sol and the RF gel, and residual RF gel. It can be seen that decomposition of carbonaceous compounds was complete when the sample was heated to 500 ◦ C. This preliminary result indicates that the RF gel induced mesoporosity in SnO2 , owing to its high thermal stability and the formation of a stable mesoporous structure. This behavior is attributed to the controlled interaction between the tin-containing sol and the RF gel, which plays an important role in the formation of the mesoporous SnO2 . Hence, this synthetic route based on an RF gel template shows potential for a variety of high-temperature applications. After calcination in the range 400–700 ◦ C the SnO2 formed in the cassiterite phase (JCPDS 41-1445). However, if the calcination temperature was lower than 500 ◦ C, we detected contamination from residual crystalline carbon (JCPDS 46-0943). Notably the products

3

Fig. 4. Nitrogen adsorption/desorption isotherms of SnO2 products prepared with (䊉) and without () the RF gel.

calcined at temperatures lower than 400 ◦ C were amorphous, and a direct phase transformation to the cassiterite phase occurred at 400 ◦ C. Representative SEM images of the Sn/RF composite before and after calcination at 500 ◦ C are shown in Fig. 3. Small grains were uniformly dispersed within the bulk of the Sn/RF composite. After calcination, the carbonaceous matrix was removed, leaving nanoparticles of SnO2 to form porous aggregates. The nitrogen adsorption/desorption isotherm, which was a Type IV isotherm with an H3-type hysteresis loop, confirmed that the calcined product was mesoporous SnO2 with slit-shaped pores (see Fig. 4). The product synthesized directly from the tin-containing sol without the RF gel showed a Type II adsorption isotherm, corresponding to multilayer adsorption on a macroporous material. These results confirmed that the RF gel is an effective template for the synthesis of mesoporous SnO2 . Effect of RF gel aging time The RF gel polycondensation process consists of two main reactions, namely an addition reaction between resorcinol and formaldehyde to form hydroxymethyl derivatives and a condensation reaction of the hydroxymethyl derivatives to form methylene and methylene ether bridged compounds. Initially, RF network clusters formed as a colloidal suspension of monomer particles. Subsequent covalent crosslinking of these particles started to occur only after many hours and then continued slowly until gelation of the RF gel occurred (Al-Muhtaseb & Ritter, 2003). To investigate the effects of gelation on the formation of mesoporous SnO2 in more detail, pre-calcined Sn/RF composites were characterized by

Fig. 3. Representative SEM micrographs of Sn/RF composite (a) before and (b) after calcination at 500 ◦ C.

Please cite this article in press as: Banjerdteerakul, K., et al. Synthesis of mesoporous tin dioxide via sol-gel process assisted by resorcinol–formaldehyde gel. Particuology (2017), https://doi.org/10.1016/j.partic.2017.07.006

G Model PARTIC-1064; No. of Pages 7 4

ARTICLE IN PRESS K. Banjerdteerakul et al. / Particuology xxx (2017) xxx–xxx Table 1 Specific surface area, pore volume, and average pore size of SnO2 products synthesized under various conditions. Pore volume (cm3 /g)

Average pore size (nm)

RF gel aging time (h) 1 28.6 2 36.8 3 38.1 4 39.6

0.029 0.047 0.051 0.058

6.5 5.2 5.4 5.8

Sn/CH2 O ratio 0.003 0.005 0.007 0.01 0.02 0.06

0.027 0.028 0.055 0.058 0.051 0.032

3.3 3.4 5.0 5.8 5.4 7.1

0.033 0.043 0.056 0.058 0.073 0.078

4.5 5.1 6.4 5.8 5.6 5.5

Condition

Fig. 5. FTIR signals corresponding to methylene bridge (), methylene ether bridge (♦), and Sn O C bonding (䊉), with respect to the aromatic ring signal for Sn/RF composites prepared from RF gels that were aged for various durations.

FTIR. The spectra were processed by Fourier self-deconvolution to enhance the resolution of the absorption bands. We determined the ratios of the signals from relevant functional groups, i.e., methylene bridge, methylene ether bridge, and Sn O C bonding represented by the bands at 1477, 1092, and 880 cm−1 , respectively, to the signal from the aromatic ring at 1614 cm−1 , for precalcined Sn/RF composites that were prepared from the RF gels that were aged for various durations. This technique eliminated the signal variation from the amount of sample analyzed. The obtained results are shown in Fig. 5. These results confirmed that the extent of both methylene and methylene ether bridge formation progressively increased as the aging time of RF gel was prolonged from 1 to 4 h. The extent of the Sn O C bonding within the Sn/RF composite also generally increased. Without aging, the RF mixture reacted violently with either the tin-precursor or the tin-containing sol, resulting in phase separation of the tin-containing liquid from the rapidly solidified RF gel. This effect prevents further reaction of the tin and the RF gel. By retarding the reactivity of the RF gel through aging, a homogenous liquid mixture of the tin-containing sol and the RF gel was achieved. This mixture featured continuous formation of the Sn O C bonding as shown in Fig. 5. The use of the aged-RF gel also increased the specific surface area of the SnO2 product. We observed that the surface area of the SnO2 product after calcination increased from 28.6 to 36.8, 38.1, and 39.6 m2 /g when the product was prepared with the RF gels that had been aged for 1, 2, 3, and 4 h, respectively (see Table 1). The increased surface area can be attributed to the RF gel structure becoming stronger and better able to support the SnO2 mesoporous structure during calcination after it had been aged for a longer time. Effect of tin-to-formaldehyde ratio in preparation of tin-containing sol To investigate the effect of the Sn/CH2 O molar ratio, all samples were prepared at an C6 H6 O2 /Na2 CO3 molar ratio of 50, while the Sn/CH2 O ratio for the tin-containing sol was varied from 0.003 to 0.06. The amount of formaldehyde used for the preparation of the RF gel was fixed such that properties of the RF gel were unchanged. The RF gel was aged for 4 h. The FTIR signal ratios presented in Fig. 6 show that the amounts of methylene and methylene ether bridges increased

Specific surface area (m2 /g)

33.0 37.2 44.0 39.6 38.1 29.5

Mixed gel aging time (day) 0 29.6 1 33.5 2 34.8 3 39.6 4 42.0 5 57.0

Fig. 6. FTIR signals corresponding to methylene bridge (), methylene ether bridge (♦), and Sn O C bonding (䊉), with respect to the aromatic ring signal for Sn/RF composites prepared from tin-containing sols of various Sn/CH2 O ratios.

as the Sn/CH2 O molar ratio was decreased. This result could be attributed to excess formaldehyde reacting with the RF gel to form more cross-links in the network of the Sn/RF composite. Notably, a high formaldehyde content in the tin-precursor/formaldehyde solution (i.e., a low Sn/CH2 O ratio) retarded the chemical activity of the tin-precursor when forming the tin-containing sol. Therefore, the tin-containing sol came into contact with the RF gel more homogeneously avoiding rapid solidification of the Sn/RF composite. This allowed for increased interactions between the tin-containing sol and the RF gel, as indicated by the enhanced signal of the Sn O C bonding. Furthermore, tin was able to diffuse into the network of the RF gel. As mentioned in Section “Effect of RF gel aging time” regarding the polycondensation of the RF gel, the addition reaction occurred during the initial step of the RF gel preparation, while the condensation reaction occurred progressively afterward. However, if the pH of the mixture was lowered during the particle growth stage, an incomplete aggregation reaction might result in more rapid condensation (Al-Muhtaseb & Ritter, 2003). Formaldehyde solution is

Please cite this article in press as: Banjerdteerakul, K., et al. Synthesis of mesoporous tin dioxide via sol-gel process assisted by resorcinol–formaldehyde gel. Particuology (2017), https://doi.org/10.1016/j.partic.2017.07.006

G Model PARTIC-1064; No. of Pages 7

ARTICLE IN PRESS K. Banjerdteerakul et al. / Particuology xxx (2017) xxx–xxx

5

Fig. 7. SEM micrographs of SnO2 products prepared with tin-containing sols with Sn/CH2 O molar ratios of 0.003 (a), 0.005 (b), 0.007 (c), 0.01 (d), 0.02 (e), and 0.06 (f).

acidic with a pH of approximately 3.0. When added to the RF gel, excess formaldehyde from the tin-containing sol at a low Sn/CH2 O ratios, disrupted growth of the RF particles owing to accelerated condensation, which affected the morphology of the SnO2 product. This behavior was confirmed from SEM images of the products obtained after calcination at 500 ◦ C, as shown in Fig. 7, where irregular aggregates were obtained when the Sn/CH2 O molar ratio was low (Fig. 7(a) and (b)). For Sn/CH2 O ratios higher than 0.005, spherical grains with a uniform size were observed, which is a typical morphology for RF gel particles. This result implies that templating of the SnO2 particles by the RF gel occurred when the condensation of the RF gel network was controlled. However, if the Sn/CH2 O ratio in the tin-containing sol was high (i.e., a high tin-precursor content), the tin-precursor reacted strongly with the RF gel resulting in rapid densification of the Sn/RF composite (see Fig. 7(f)). This result is supported by nitrogen adsorption/desorption isotherms

of the calcined products presented in Fig. 8(a), where the product prepared using a high Sn/CH2 O ratio showed a Type II adsorption isotherm, similar to that of the product prepared without the RF gel. The pore volumes of samples shown in Fig. 7(a) and (b) were in the range 0.027–0.028 cm3 /g, while the average pore diameters were in the range 3.3–3.4 nm (see Table 1). As the Sn/CH2 O molar ratio decreased, the reaction proceeded more slowly, allowing the tin to disperse into mesopores of the RF network, which had already taken on a spherical shape. Hence, when the RF gel was removed by calcination, the SnO2 product retained its spherical shape and mesoporosity as indicated by the shift of the nitrogen adsorption/desorption isotherm to a Type IV and the pore-size distribution, as shown in Fig. 8(b). The pore volume and average pore diameter also increased to 0.051–0.058 cm3 /g and 5.0–5.8 nm, respectively (see Table 1). When the Sn/CH2 O ratio was 0.06, only a small fraction of the tin could access the mesopores of the RF

Please cite this article in press as: Banjerdteerakul, K., et al. Synthesis of mesoporous tin dioxide via sol-gel process assisted by resorcinol–formaldehyde gel. Particuology (2017), https://doi.org/10.1016/j.partic.2017.07.006

G Model PARTIC-1064; No. of Pages 7

ARTICLE IN PRESS K. Banjerdteerakul et al. / Particuology xxx (2017) xxx–xxx

6

Fig. 8. Nitrogen adsorption/desorption isotherms (a) and pore-size distribution (b) of SnO2 products prepared with tin-containing sols with various Sn/CH2 O molar ratios.

network because of the rapid densification of the mixture. Consequently, the majority of tin-containing sol aggregates formed without the templating effect, resulted in a partial shift of the isotherm toward a Type II isotherm and disappearance of the mesoporosity. The pore volume decreased to 0.032 cm3 /g, while the average pore size widened to 7.1 nm (see Table 1). Notably, the specific surface area values of the SnO2 products synthesized at Sn/CH2 O ratios of 0.003, 0.005, 0.007, 0.01, 0.02, and 0.06 were 33.0, 37.2, 44.0, 39.6, 38.1, and 29.5 m2 /g, respectively (see Table 1). Effect of resorcinol-to-carbonate molar ratio Sodium carbonate was used to catalyze the formation of the hydroxymethyl derivatives of resorcinol prior to condensation of these derivatives to form the bridged RF network. In this study, the C6 H6 O2 /Na2 CO3 molar ratio for preparation of the RF gel was varied from 50 to 300, while the C6 H6 O2 /CH2 O molar ratio was maintained at 0.5. These C6 H6 O2 /Na2 CO3 ratios were considered to be in the typical range used for RF gel synthesis (Al-Muhtaseb & Ritter, 2003). The Sn/RF composites were prepared with a tin-containing sol synthesized with a Sn/CH2 O molar ratio of 0.01. The composites were analyzed by FTIR in the same manner as previously described. We found that the signal ratios corresponding to the methylene bridge, methylene ether bridge, and Sn O C bonding varied only slightly without an obvious trend in the ranges of 1.18–1.21, 0.79–0.80, and 0.26–0.31, respectively (results not shown). These results indicate that the variation of functional groups in the Sn/RF composite was not markedly influenced by the amount of catalyst used in the considered range. We found that as the C6 H6 O2 /Na2 CO3 molar ratio was increased from 50 to 300, the specific surface area of the calcined SnO2 product decreased from 39.6 to 31.2 m2 /g. Because a high C6 H6 O2 /Na2 CO3 ratio resulted in an RF colloidal gel consisting of large RF particles with large interparticle spacing (Al-Muhtaseb & Ritter, 2003), these results suggested that the tincontaining sols filled the voids between RF particles and grew into large SnO2 grains during the calcination, leading to a decrease of the surface area at high C6 H6 O2 /Na2 CO3 molar ratios.

Fig. 9. FTIR signal ratios of methylene bridges (), methylene ether bridges (♦), and Sn O C bonding (䊉) to the aromatic ring signal for Sn/RF composites with various Sn/RF mixed gel aging durations, compared with the corresponding variation of specific surface area (䊏) of the SnO2 products obtained after calcination of the Sn/RF composites at 500 ◦ C for 6 h.

tin-containing sol and the RF template might be more effective for preventing pore collapse during the calcination. A SnO2 product achieved the highest specific surface area (57.0 m2 /g) among the products synthesized in this work (see Table 1). Hence, sufficient time for the RF network to develop into a porous network is an important factor. As discussed in the previous sections, although some conditions resulted in extensive crosslinking within the RF network, an uncontrolled interaction between the tin-precursor and RF gel, causing the composite to solidify rapidly, does not result in the desired porous structure.

Conclusions Effect of Sn/RF mixed gel aging time The effect of aging the Sn/RF mixed gel are summarized in Fig. 9. All samples were synthesized from a tin-containing sol with a Sn/CH2 O ratio of 0.01 mixed with an RF gel that was prepared with a C6 H6 O2 /Na2 CO3 ratio of 50, aged for 4 h. These results show that the condensation reaction of the hydroxymethyl derivatives of resorcinol to form methylene and methylene ether bridges progressed continuously for approximately 3 days after the RF gel was mixed with the tin-containing sol. The reaction between the RF gel and the tin-containing sol continued to form Sn O C bonds. The strengthened network of the RF gel and the strong interactions between the

Mesoporous SnO2 was successfully synthesized via a sol–gel method with an RF gel as a template. The strong interactions between the RF gel and tin-containing sol influenced the structural strength of the resulting mesoporous network. The morphology and mesoporosity of the final product can be effectively controlled by varying the synthesis conditions (i.e., aging time of the RF gel, Sn/CH2 O molar ratio, C6 H6 O2 /Na2 CO3 molar ratio, and aging time of the Sn/RF mixed gel), resulting in homogeneous diffusion of the tin-containing sol into the network of the RF gel during the condensation reaction. As a consequence, the resulting mesoporous tin dioxides exhibited satisfactorily high surface areas up to 57 m2 /g.

Please cite this article in press as: Banjerdteerakul, K., et al. Synthesis of mesoporous tin dioxide via sol-gel process assisted by resorcinol–formaldehyde gel. Particuology (2017), https://doi.org/10.1016/j.partic.2017.07.006

G Model PARTIC-1064; No. of Pages 7

ARTICLE IN PRESS K. Banjerdteerakul et al. / Particuology xxx (2017) xxx–xxx

Acknowledgement This work was partially supported by Thailand Research Fund (TRF). References Ahmed, J., Vaidya, S., Ahmad, T., Devi, P. S., Das, D., & Ganguli, A. K. (2008). Tin dioxide nanoparticles: Reverse micellar synthesis and gas sensing properties. Materials Research Bulletin, 43, 264–271. Ahmad, M. Y., Abu Bakar, M. Z., & Jusliha, J. (2016). Synthesis of mesoporous tin oxide with cationic surfactant as a template in aqueous media. American Journal of Chemistry, 6(5), 145–147. Al-Muhtaseb, S. A., & Ritter, J. A. (2003). Preparation and properties of resorcinol–formaldehyde organic and carbon gels. Advanced Materials, 15, 101–114. Babic, B., Kaluderovic, B., Vracar, L., & Krstajic, N. (2004). Characterization of carbon cryogel synthesized by sol–gel polycondensation and freeze-drying. Carbon, 42, 2617–2624. Bruno, L., Pijolat, C., & Lalauze, R. (1994). Tin dioxide thin-film gas sensor prepared by chemical-vapor-deposition: Influence of grain-size and thickness on the electrical-properties. Sensors and Actuators B: Chemical, 18, 195–199. Chantam, P., & Pavarajarn, V. (2012). Synthesis of porous alumina assisted by resorcinol–formaldehyde gel. Journal of the American Ceramic Society, 95, 2788–2790. Chen, F. L., & Liu, M. L. (1999). Preparation of mesoporous tin oxide for electrochemical applications. Chemical Communications, 1829–1830. Emons, T. T., Li, J. Q., & Nazar, L. F. (2002). Synthesis and characterization of mesoporous indium tin oxide possessing an electronically conductive framework. Journal of the American Chemical Society, 124, 8516–8517. Fujihara, S., Maeda, T., Ohgi, H., Hosono, E., Imai, H., & Kim, S. H. (2004). Hydrothermal routes to prepare nanocrystalline mesoporous SnO2 having high thermal stability. Langmuir, 20, 6476–6481. Guo, X., Kang, Y., Wang, L., Liu, X., Zhang, J., Yang, T., et al. (2011). Mesoporous tin dioxide nanopowders based sensors to selectively detect ethanol vapor. Materials Science and Engineering C, 31, 1369–1373. Hieda, K., Hyodo, T., Shimizu, Y., & Egashira, M. (2008). Preparation of porous tin dioxide powder by ultrasonic spray pyrolysis and their application to sensor materials. Sensors and Actuators B: Chemical, 133, 144–150. Hyodo, T., Nishida, N., Shimizu, Y., & Egashira, M. (2001). Preparation of thermally stable mesoporous tin dioxide powders with high specific surface area by utilizing self-assembly of surfactants. Journal of the Ceramic Society of Japan, 109, 481–483. Jain, K., Rashmi, & Lakshmikumar, S. T. (2005). Preparation of nanocrystalline tin oxide powder for gas sensor applications. Journal of Surface Science and Technology, 21, 129–138. Jin, Z. H., Zhou, H. J., Jin, Z. L., Savinell, R. F., & Liu, C. C. (1998). Application of nanocrystalline porous tin oxide thin film for CO sensing. Sensors and Actuators B: Chemical, 52, 188–194.

7

Jirglova, H., & Maldonado-Hodar, F. J. (2010). Chemical interactions of surface-active agents with growing resorcinol–formaldehyde gels. Langmuir, 26, 16103–16109. Krissanasaeranee, M., Supaphol, P., & Wongkasemjit, S. (2010). Preparation of poly(vinyl alcohol)/tin glycolate composite fibers by combined sol–gel/electrospinning techniques and their conversion to ultrafine tin oxide fibers. Materials Chemistry and Physics, 119, 175–181. Li, L. L., Zhang, W. M., Yuan, Q., Li, Z. X., Fang, C. J., Sun, L. D., et al. (2008). Room temperature ionic liquids assisted green synthesis of nanocrystalline porous SnO2 and their gas sensor behaviors. Crystal Growth & Design, 8, 4165–4172. Lin, C., & Ritter, J. A. (1997). Effect of synthesis pH on the structure of carbon xerogels. Carbon, 35, 1271–1278. Luyjew, K., Tonanon, N., & Pavarajarn, V. (2008). Mesoporous silicon nitride synthesis via the carbothermal reduction and nitridation of carbonized silica/RF gel composite. Journal of the American Ceramic Society, 91, 1365–1368. Pekala, R. W. (1989). Organic aerogels from the polycondensation of resorcinol with formaldehyde. Journal of Materials Science, 24, 3221–3227. Poljansek, I., & Krajnc, M. (2005). Characterization of phenol-formaldehyde prepolymer resins by in line FT-IR spectroscopy. Acta Chimica Slovenica, 52, 238–244. Shimizu, Y., Jono, A., Hyodo, T., & Egashira, M. (2005). Preparation of large mesoporous SnO2 powder for gas sensor application. Sensors and Actuators B: Chemical, 108, 56–61. Song, K. C., & Kim, J. H. (2000). Synthesis of high surface area tin oxide powders via water-in-oil microemulsions. Powder Technology, 107, 268–272. Toupance, T., Babot, O., Jousseaume, B., & Vilaca, G. (2003). Nanocrystalline mesoporous tin dioxide prepared by the sol-gel route from a dialkoxydi(betadiketonato)tin complex. Chemistry of Materials, 15, 4691–4697. Wang, Y., Ma, C., Sun, X., & Li, H. (2001). Synthesis of mesoporous structured material based on tin oxide. Microporous and Mesoporous Materials, 49, 171–178. Wen, Z. H., Wang, Q., Zhang, Q., & Li, J. H. (2007). In situ growth of mesoporous SnO2 on multiwalled carbon nanotubes: A novel composite with porous-tube structure as anode for lithium batteries. Advanced Functional Materials, 17, 2772–2778. Xi, G., He, Y., Zhang, Q., Xiao, H., Wang, X., & Wang, C. (2008). Synthesis of crystalline microporous SnO2 via a surfactant-assisted microwave heating method: A general and rapid method for the synthesis of metal oxide nanostructures. Journal of Physical Chemistry C, 112, 11645–11649. Yu, A. S., & Frech, R. (2002). Mesoporous tin oxides as lithium intercalation anode materials. Journal of Power Sources, 104, 97–100. Zhang, P., Huang, S. Y., & Popov, B. N. (2010). Mesoporous tin oxide as an oxidationresistant catalyst support for proton exchange membrane fuel cells. Journal of the Electrochemical Society, 157, B1163–B1172. Zhou, S. X., Lu, S. M., Ke, Y. X., & Li, J. M. (2003). Synthesis of mesostructured tin oxide with neutral surfactant as a template in aqueous media. Materials Letters, 57, 2679–2681.

Please cite this article in press as: Banjerdteerakul, K., et al. Synthesis of mesoporous tin dioxide via sol-gel process assisted by resorcinol–formaldehyde gel. Particuology (2017), https://doi.org/10.1016/j.partic.2017.07.006