J Sol-Gel Sci Techn (2007) 41:139–146 DOI 10.1007/s10971-006-0513-y
Effective preparation of crack-free silica aerogels via ambient drying Sung-Woo Hwang · Hae-Hyun Jung · Sang-Hoon Hyun · Young-Soo Ahn
Received: 1 February 2006 / Accepted: 19 June 2006 / Published online: 3 January 2007 C Springer Science + Business Media, LLC 2007
Abstract Effective ambient-drying techniques for synthesizing crack-free silica aerogel bulks from the industrial waterglass have been developed. Silica wet gels were obtained from aqueous colloidal silica sols prepared by ion-exchange of waterglass solution (4–10 wt% SiO2 ). Crack-free monolithic silica aerogel disks (diameter of 22 mm and thickness of 7 mm) were produced via solvent exchange/surface modification of the wet gels using isopropanol/trimethylchlorosilane/n-Hexane solution, followed by ambient drying. The effects of the silica content in sol and the molar ratio of trimethylchlorosilane/pore water on the morphology and property of final aerogel products were also investigated. The porosity, density, and specific surface area of silica aerogels were in the range of 92–94%, 0.13–0.16 g/cm3 , and ∼ 675 m2 /g, respectively. The degree of springback during the ambient drying processing of modified silica gels was 94%. Keywords Aerogel . Silica . Surface modification . Ambient drying
1 Introduction The extraordinary properties of silica aerogels including high porosity ( ∼ 99%), high specific surface area S.-W. Hwang · H.-H. Jung · S.-H. Hyun () School of Advanced Materials Science and Engineering, Yonsei University, 134 Shinchon-dong, Seodaemun-Gu, Seoul 120-749, Korea e-mail:
[email protected] Y.-S. Ahn korea Institute of Energy Research, Daejeon 62-1, Korea
(600–1200 m2 /g), low density (0.003–0.15 g/cm3 ), and low thermal conductivity (0.01–0.03 W/m · K) lead to numerous potential applications such as thermal insulating and catalyst materials [1–3]. Because of their potential for wide industrial application, silica aerogels have received significant attentions [2]. However, conventional synthesis of silica aerogel have been limited by expensive starting materials like alkoxides and supercritical drying process which is not only expensive but also dangerous and difficult to industrialization. For commercialization of silica aerogel, the reduction of costs and risks has been investigated. The research took the following two approaches: (1) the use of inexpensive starting materials such as waterglass (sodium silicate) and (2) the development of ambient pressure drying techniques instead of conventional supercritical drying [4–8]. The ambient pressure drying is possible through the solvent exchange and surface modification of the wet gels. Surface silanol groups that lead the gel to collapse by condensation were modified into the non-reactive organic radicals. Recently, Schwertfeger et al. [9] reported a process for waterglass based silica aerogel powders and pieces in which solvent exchange and surface modification using mixed solution of organic solvent and chlorosilane were simultaneously progressed by chemical reactions and a phase separation mechanism. Furthermore, approaches for producing of silica aerogels by ambient pressure drying using various solvent exchange/surface modification agents were gradually reported [10–12]. In this work, a newly modified and effective techniques for synthesizing crack-free silica aerogel bulks via ambient pressure drying, which uses isopropanol (IPA)/trimethylchrolosilane (TMCS)/n-Hexane solution as a solvent exchange/surface modification agent, were explored. As a starting material, inexpensive industrial waterglass solution was selected and the effective ion-exchanging system Springer
140
J Sol-Gel Sci Techn (2007) 41:139–146
Industrial waterglass ( Na2O·3.3SiO2·22.6H2O) Addition of D.I water
Stirring
Waterglass solution (4~10 wt%) Ion-exchange with stirring Sodium-free silica sol (Silicic Acid ) pH Adjustment
NH4OH (1M)
Wet gel formation Aging in D. I water at 60oC Solvent exchange / Surface modification in IPA/TMCS/n-Hexane Ambient drying Silica aerogel
Fig. 1 Overall flowchart for synthesis of silica aerogel by IPA/TMCS/n-Hexane method and ambient pressure drying
for producing silica sol was designed. The effects of the process variables on the morphologies and the properties of silica aerogel bulks are also discussed. 2 Experimental Procedure The overall experimental flowchart and schematic representation for synthetic procedure of crack-free silica aerogel bulks are described in Figs. 1 and 2, respectively. 2.1 Preparation of colloidal silica sol and wet gels Silica sols used in this work were prepared with waterglass (sodium silicate) solution (Na2 O:SiO2 molar ratio of 1:3.3, Fig. 2 Schematic representation of overall synthetic process for aerogels by ambient pressure drying
Springer
Il-shin Chem., Korea) as a starting material. A strong acidic cation resin (Amberlite IR-120H, Rohm and Hass, France) of the sulphonated polystyrene type was selected to remove sodium ions in the waterglass. For an effective ion-exchange between the sodium ion in waterglass and the proton in the resin, stirring-fluidized bed type of the ion-exchanging system was manufactured as shown in Fig. 3. This system was designed to realize continuous-fluidized condition in the column and to minimize the silica loss during the ion-exchange process. About 0.75 liter of waterglass solution having a SiO2 contents of 4 to 10 wt% went through an ion-exchange column filled with 1 liter ( = 800 g) of ion-exchange resin. The waterglass solution was flowed into the ion-exchange column at a rate of 60 ml/min, and the stirring speed was fixed at 600 rpm. During stirring, the ion-exchanged waterglass solution (silica sol) was collected at a rate of 60 ml/min. The collected aqueous colloidal silica sol had a pH in the range of 2.4–2.7. For gelation, the initial pH of the silica sol was adjusted to 3.5 with diluted ammonia water. Then, silica sol having a pH value of 3.5 was decanted into the polypropylene cylindrical vessels. The vessels containing silica sol were sealed with paraffin film. After 40 min of gelation at 60◦ C, the silica wet gels (hydrogel) were aged for 1 day in deionized water in order to strengthen the network structure of the gels. 2.2 Solvent exchange/surface modification and ambient pressure drying of wet gels Three solvents, isopropyl alcohol (IPA; Yakuri pure chem., Japan), trimethylchlorosilane (TMCS; Lancaster, UK), and n-Hexane (Duksan pure chem., Korea) were used for the solvent exchange/surface modification process. The aged silica wet gels containing water in their internal pores were immersed in IPA/TMCS/n-Hexane solution, and the conditions for solvent exchange/surface modification were controlled as follows: the molar ratio of IPA:pore water:TMCS = 0.25–0.4:1:0.25–0.4 and the volumetric ratio of n-Hexane/TMCS = 10. After 1 day of the modification procedure at 60◦ C, the surface-modified wet gels were dried at room temperature by controlled evaporation of pore liquid in the n-Hexane atmosphere for 3 days. And then, the dried
J Sol-Gel Sci Techn (2007) 41:139–146
141
Fig. 3 Stirring-fluidized bed type ion-exchanging system for producing silica sol used in this work
gels were heat-treated at 50◦ C for 1 h and 230◦ C for 1 h in air. 2.3 Characterization The residual sodium ion amount in silica sols was measured by using the sodium ion detector (pNa 205-1000S7 106375, WTW, Germany) and the inductively coupled plasma–atomic emission spectrometer (ICP-AES, Varian SpectrrAA800, Australia). The silica content in silica sols was determined by weighing solids after the heat treatment of silica sols at 750◦ C for 2 h in a box furnace. The bulk density of the silica wet gels and aerogels was determined by weighing samples of known dimensions, and the porosity was calculated from the Eq. (1): Porosity = (1 − ρ/ρs ) × 100(%),
(1)
where ρ is the density of aerogel and ρs is the theoretical density of thermal SiO2 . Springback efficiency during ambient pressure drying was simply calculated by the Eq. (2): Springback efficiency = (Va /Vw ) × 100(%),
(2)
where Va is the bulk volume of silica aerogel and Vw is the volume of wet gel before modification. The specific surface area of silica aerogel was measured using the Brunauer-Emmitt-Teller (BET) method (Model Gemini 2375, Micromeritics Instrument Group, Norcross, GA).
3 Results 3.1 Properties of silica sol During the gelation and aging procedures, syneresis of wet gels was observed and the linear shrinkage of wet gel caused by syneresis was determined to be about 10%. The optimum pH, gelation temperature, and aging time were 3.5 pH, 60◦ C, and 24 h, respectively, from the results of our previous work [8]. Figure 4 shows the SiO2 content of produced silica sols after ion-exchange. The content is dependent upon the initial silica amounts of the starting waterglass solution. As the initial silica content of the waterglass solution was increased, the SiO2 content of produced silica sol was also increased during the ion-exchange procedure. The silica loss during
Springer
142
J Sol-Gel Sci Techn (2007) 41:139–146
Fig. 4 Variation of SiO2 content in silica sols after ion-exchange depending on the initial SiO2 amounts in the starting waterglass solution
10
No SiO2 loss SiO2 content after ion-exchange
SiO2 content of silica sols (wt%)
8
6
4
2
0 0
2
4
6
8
10
Initial SiO 2 content in waterglass (wt%)
the ion-exchange process using stirring-fluidized bed type system can be described by Eq. (3), from Fig. 4. Cs = 0.81532 + 0.83098Cw
(3)
where Cs is the silica content in silica sol and Cw is the initial silica amount of starting waterglass solution; so, aqueous silica sols with the specific SiO2 content can be freely obtained from waterglass solution in the range of 4 to 10 wt% SiO2 . Table 1 shows the amounts of residual sodium ion in silica sols after the ion-exchange procedure, according to initial SiO2 content in starting waterglass solution. Independent of the initial SiO2 content in starting waterglass solution, the amounts of sodium ions in silica sol were detected below 1 ppm by the sodium ion detector and the ICP-AES. Table 1 Amounts of residual sodium ion in silica sol after ionexchange procedure a
Initial SiO2 contents in waterglass solution (wt%)
Residual sodium ion after ionexchange (ppm)
pH of sol
SiO2 contents in silica sols (wt%)
4 6 6.25 6.3 8 10
0.398 0.928 0.828 0.404 0.196 0.313
2.45 2.37 2.27 2.24 2.19 2.16
3.62 5.61 5.92 6.14 7.68 9.01
b
Gelation time (minutes)
58.1 26.5 24.6 18.1 11.1 8.4
a
Averaged value.
b
Gelation conditions: pH value of silica sol = 3.5 pH and at 60◦ C.
Springer
Fig. 5 TEM image of silica sol during gelation (SiO2 content of silica sol = 6 wt%)
In the TEM micrograph of silica sol (SiO2 content of 6 wt%) in Fig. 5, the colloidal silica particles having the size of 70 to 80 nm in diameter were observed. Figure 5 also shows the formation of the silica network structures during the gelation step of silica sol. 3.2 Properties of surface-modified gels and aerogels The ‘springback’ effect during ambient pressure drying process after surface modification is shown in Fig. 6. After solvent exchange/surface modification, modified wet gels were dried in n-Hexane atmosphere. The linear shrinkage during this stage was measured up to 50%. However, after further drying and heat treatment, the shrunken gels began to expand gradually, and their size and volume were 95% restored to their original modified gels state.
J Sol-Gel Sci Techn (2007) 41:139–146
Fig. 6 Springback phenomena during ambient pressure drying process
Figure 7 shows the effects of the SiO2 content and the TMCS/pore water molar ratios on the morphologies of aerogels after ambient pressure drying. The surface-modified wet gels having SiO2 content of less than 4 wt% collapsed during drying. The disk-type bulk shape of the surface-modified gels was maintained without visible cracks, where the SiO2 content of wet gels ranged from 4 to 8 wt%. However, in
143
the case where the SiO2 content was over 8 wt%, many visible cracks were observed on the surface-modified wet gels during drying as shown in Fig. 7(a). The effect of TMCS/pore water molar ratio on the bulk shape of the aerogel is shown in Fig. 7(b). When the TMCS/pore water molar ratio was less than 0.25, the surfacemodified wet gels showed a yellowish color and the springback phenomena did not occur during drying. In contrast to the above result, where the TMCS/pore water ratio was more than 0.45, the modified wet gels were disrupted by crack generation during drying. In the range of TMCS/pore water molar ratio of 0.297 to 0.396, the modified wet gels were dried without any visible cracks, and complete springback phenomena was also observed. After ambient pressure drying and heat treatment, crack-free and translucent silica aerogels having diameters of 22.5 mm to 23 mm and thicknesses of 6.5 mm to 7.4 mm were successfully obtained. Figure 8 shows bulk density and porosity of synthesized aerogels according to TMCS/pore water molar ratios. The bulk densities of silica aerogels were varied in the range
Fig. 7 Effects of SiO2 content and TMCS/pore water molar ratio on the morphologies of aerogels after ambient pressure drying Springer
144
J Sol-Gel Sci Techn (2007) 41:139–146
Fig. 8 Plot of bulk density and porosity of silica aerogels depending on TMCS/pore water molar ratios
0.40
100 98 96
3
Bulk density (g/cm )
0.35
0.30
0.25
92
0.20
90
Porosity (%)
94
88 0.15
0.10 0.00 0.24
0 0.26
0.28
0.30
0.32
0.34
0.36
0.38
0.40
Molar ratio of TMCS/pore water structure with continuous porosity and coagulated particle size less than 100 nm were maintained. This was caused by the springback effect during ambient pressure drying process.
4 Discussion From Figs. 4 to 5 and Table 1, it was determined that the stirring-fluidized bed type ion-exchanging system used in this work is very effective for producing sodium-free aqueous colloidal silica sol from the waterglass solution. Because the continuous-fluidized condition, in the ion-exchanging 100
650
98
600
96
550
94
2
700
500
92
450
90
400 88 86
50
BET surf ace area Springback efficiency 0 0.24
0 0.26
0.28
0.30
0.32
0.34
0.36
Molar rario of TMCS/pore water Springer
0.38
0.40
Springback efficiency (%)
Fig. 9 Variations of BET surface area of silica aerogels and springback efficiencies during ambient pressure drying depending on various TMCS/pore water molar ratios
Specific surface area ( m /g )
of 0.135 g/cm3 to 0.167 g/cm3 as the TMCS/pore water ratio increased. And the porosities of silica aerogels were determined in the range of 92.4% to 93.9% depending on the TMCS/pore water molar ratio. The variations of specific surface area and springback efficiency during ambient drying, depending on the TMCS/pore water molar ratio, are given in Fig. 9. The highest specific surface area (675 m2 /g) of silica aerogel was obtained from the gels modified with TMCS/pore water molar ratio of 0.396. The springback efficiency was determined up to 94.5% as the TMCS/pore water ratio increased. Figure 10 shows a microstructure of silica aerogel synthesized by ambient pressure drying. The highly porous network
J Sol-Gel Sci Techn (2007) 41:139–146
Fig. 10 SEM micrograph of silica aerogel (TMCS/pore water molar ratio of 0.35, bulk density of 0.135 g/cm3 , porosity of 93.8%, and specific surface area of 662 m2 /g)
column, prevented the gelation and agglomeration of silica particles between the resin particles, silica sols having the desired SiO2 contents could be freely synthesized with very low silica loss during ion exchange process. Therefore, the ion-exchanging concept in Fig. 3 could be a suitable design for industrial application because the minimization of silica loss and complete removal of sodium ions are possible. During the solvent exchange/surface modification process, using IPA/TMCS/n-Hexane solution, a light-yellowish liquid seeped out of the wet gel and formed under the nHexane phase by phase separation. The resulting surfacemodified wet gel was located on the interface of product liquid and n-Hexane. These phenomena could be determined by the several reactions that can occur during the solvent exchange/surface modification of wet gel. The basic concept of this solvent exchange/surface modification was already presented in previous work [7]. The proposed mechanism of solvent exchange/surface modification of wet gels in the IPA/TMCS/n-Hexane solution
145
is described in Fig. 11. TMCS reacts with pore water (H2 O) and IPA. And then, hexamethyldisiloxane (HMDSO), isopropoxytrimethylsilane [(CH3 )3 Si-O-CH(CH3 )2 ], and HCl were spontaneously generated in the pores of wet gel. Isopropoxytrimethylsilane and TMCS react with the surface silanol group of SiO2 network causing the surface modification. According to these surface modification reactions, the initial hydrophilic character of the internal surface of SiO2 network was transformed into hydrophobic. Consequently, hydrophobic areas were formed on the surface of the SiO2 network structures. The hydrophobic areas and adjacent layer grew into the gel while the reactions proceeded. Finally, aqueous HCl/IPA solution that separated from the main phase (residual IPA/n-Hexane solution) in the pores of wet gels seeped out from the pores of the surface-modified wet gel. So, phase separation between main phase and aqueous HCl/IPA solution occurred outside of the wet gel. Because of aqueous HCl/IPA solution is denser than n-Hexane, it is located at the bottom of the main phase. Therefore, the surface-modified wet gel filled with the n-Hexane as a pore liquid is located on the aqueous HCl/IPA phase after complete surface modification. For effective surface modification of the wet gel, experimental investigation of the optimum TMCS/pore water molar ratio is most important in this work, especially. In case of the TMCS/pore water molar ratio less than 0.25, the surface silanol groups on the SiO2 network structures of the wet gel did not fully react with the modification agents. Because of insufficient modification, the generated HCl/IPA solution can not seep out from the pores of wet gel. These partially unmodified wet gels did now show the springback phenomena during drying as given in Fig. 6, and they had a light-yellowish color caused by HCl trapped in the pores of the gel. However, many cracks were generated on the gels modified with the IPA/TMCS/n-Hexane solution having a TMCS/pore water molar ratio of over 0.4, during drying
Fig. 11 Proposed mechanism of solvent exchange/surface modification by IPA/TMCS/n-Hexane method developed in this work Springer
146
process. In this case, it can be thought that the excess modification and presence of remaining TMCS causes internal cracking of the SiO2 network during the modification process [10]. These cracks caused a fracture of the disk-type bulk structure during springback. The reasons for cracking during modification process were already reported [9–13], but more scientific research is still needed to clarify the cracking mechanism. In this work, a monolithic disk of silica aerogels without cracks could be obtained when the TMCS/pore water molar ratio was 0.297 to 0.396, as shown in Fig 7. The bulk density, porosity, specific surface area, and springback efficiency were almost constant in the above samples as given in Figs. 8 and 9. But, it is still difficult to produce crackfree aerogel bulks with a higher yield than 50% by ambient pressure drying, because of the irreversible shrinkage and micro-cracking generated by capillary pressure differences. To achieve the high production yield of crack-free aerogel bulks, more detailed research into the prevention of cracking and optimization of ambient-drying condition are required.
5 Conclusion The stirring-fluidized ion-exchanging system was designed for the effective producing of aqueous colloidal silica sols using waterglass solution as a starting material. Using this system, the silica loss during ion exchange and sodium ions in silica sol are minimized to less than 10% and 1 ppm, respectively. Crack-free silica aerogel bulks (disk type, diameter of ∼ 22.5 mm and thickness of ∼ 0.7 mm) have
Springer
J Sol-Gel Sci Techn (2007) 41:139–146
been synthesized via instantaneous solvent exchange/surface modification using IPA/TMCS/n-Hexane solution and ambient pressure drying techniques. The TMCS/pore water molar ratio for effective solvent exchange/surface modification was optimized in the range of 0.297 to 0.396. The resulting crack-free silica aerogel bulk have a hydrophobic surface, which shows a high specific surface area ( ∼ 675 m2 /g), low density ( ∼ 0.13 g/cm3 ), high porosity ( ∼ 94%), and high springback efficiency ( ∼ 94.5%). Acknowledgment This work was financially supported by the Ministry of Commerce, Industry and Energy, Korea Energy Management Corporation through the Energy Conservation Technology R&D program.
References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13.
Fricke J, Tillotson T (1997) Thin Solid Films 297:212 Fricke J, Emmerling A (1998) J Sol-Gel Sci Tech 13:299 Pierre AC, Pajonk GM (2002) Chem Rev 102:4243 Herrmann G, Iden R, Mielke M (1995) J Non-Cryst Solids 186:380 Tyler LJ (1962) Dow Corning, U.S. Patent 3,015,645 Jansen M, Zimmermann A (1997) Hoechst, U.S. Patent 5,647,962 Lee CJ, Kim GS, Hyun SH (2002) J Mater Sci 37:2237 Kim GS, Hyun SH (2003) J Mater Sci 38:1961 Schwertfeger F, Frank D, Schmit M (1998) J Non-Cryst Solids 225:24 Venkateswara Rao A, Nilsen E, Enarsrud M-A (2001) J Non-Cryst Solids 296:165 Venkateswara Rao A, Pajonk GM, Bhagat SD, Barboux P (2004) J Non-Cryst Solids 350:216 Shi Fei, Wang Lijiu, Liu Jingxiao (2006) Mater Lett 60: 3718 Field RJ, Olson EW (2001) J Non-Cryst Solids 285:194
