aerogel

j o u r n a l o f m a t e r i a l s p r o c e s s i n g t e c h n o l o g y 1 9 9 ( 2 0 0 8 ) 10 10–26

  journal homepage: www.elsevier.com/locate/jmatprotec

Review

Silica aerogel; synthesis, properties and characterization  A. Soleimani Dorcheh , M.H. Abbasi ∗

Department of Materials Engineering, Isfahan University of Technology, Isfahan, Iran

a rt ic le

i nf o

 Article history: Received 5 September 2007 Accepted 23 October 2007

Keywords: Silica aerogel Synthesis Characterization Nanostructure

a b s t r a c t In recent years, silica aerogels have attracted increasingly more attention due to their extraordinary properties and their existing and potential applications in wide variety technological areas. Silica aerogel is a nanostructured material with high specific surface area, high porosity, low density, low dielectric constant and excellent heat insulation properties. Many research works have been carried out concerning aerogel production and characterization. In this review paper, research work and developments in synthesis, properties and characterization of silica aerogels will be addressed. Particular attention is paid to drying which is a critical step in aerogel synthesis and makes the production of this material more economical and commercial. © 2007 Elsevier B.V. B.V. All rights reserved.

Contents 1. 2.

3.

4.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Gel preparation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.1. Starting materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.2. Sol preparation and gelation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. Additives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3. Aging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4. Drying . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4.1. Supercritical drying (SCD) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4.2. Ambient pressure drying (APD) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4.3. Freeze drying . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Properties of silica aerogels and methods of determination. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. . 3.1. Pore structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. Density . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3. Optical properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4. Thermal conductivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5. Hydrophobicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Corresponding author . Tel.: +98 312 376 0908. E-mail address: ali [email protected] [email protected] r (A. Soleimani Dorcheh). 0924-0136/$ – see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.jmatprotec.2007.10.060 doi:10.1016/j.jmatprotec.2007.10.060 ∗

11 11 11 11 12 14 14 14 15 15 16 18 21 21 21 22 22 23 23 24 24

  journal of materials processing technology 199 (2008) 10–26

1.

Introduction

Silica aerogels are materials with unusual properties such as high specific surface area (500–1200 m 2  /g), high porosity (80–99.8%), low density (∼0.003 g/cm3 ), high thermal insulation value (0.005W/mK), ultra low dielectric constant (k =1.0–2.0) and low index of refraction ( ∼1.05) (Fricke, 1988; Fricke and Emmerling, 1992; Kim and Hyun, 2001; Carraher, 2005; Schultz et al., 2005a). Silica aerogels were first produced in the 1930s, but saw little development for several decades. The revival of sol–gel technology is owed, in large part, to the attempts of Stanislaus Teichner at Universite Claud Bernard, to produce silica aerogel for the storage of rocket fuels ( Pajonk and Teichner, 1985). Since then the proposed applications of aerogel have been many and varied, yet most remain unrealized. Commercial applications such as thermal window insulation, acoustic barriers, supercapacitors and catalytic supports have all been proposed, but little in the way of actual use has resulted (Hrubesh, 1998; Schmidt and Schwertfeger, 1998; Ulrich, 1990; Fricke and Tillotson, 1997; MacKenzie, 1988; Pierre and Pajonk, 2002). However, monolithicsilica aerogel has been usedextensively in high energy physics in Cherenkov radiation detectors (Adachi et al., 1995; Asner et al., 1996; Sumiyoshi et al., 1998; Ishino et al., 2001; DeLeo et al., 1997 ). Other application of silica aerogels are: shock wave studies at high pressures ( Tewari et al., 1986), inertial confinement fusion (ICF) (Anappara et al., 2004), radioluminescent devices (Rao et al., 2003) and micrometeorites (Nyquist and Kagel, 1997). The synthesis of silica aerogels has received significant attention especially during the last two decades. Some investigators have studied the use of different precursors and many have focused on modification of synthesis parameters (Schwertfeger et al., 1998; Tyler, 1959; Tyler, 1962). Recently many research works have been devoted to ambient pressure drying which makes the production commercial and industrial (Schwertfeger et al., 1998; Lee et al., 2002; Parvathy Rao et al., 2005). This paper reviewed the research work carried out on synthesis and characterization of silica aerogels during the past two decades.

2.

Synthesis

The synthesis of silica aerogels can be divided into 3 general steps: (a) Gel preparation . The silica gel is obtained by sol–gel process. The sol is prepared by a silica source solution and by addition of catalyst, gelation is occurred. The gels are usually classified according to the dispersion medium used, e.g., hydrogel or aquagel, alcogel and aerogel (for water, alcohol, and air, respectively). (b) Aging of the gel . The gel prepared in the first step is aged in its mother solution. This aging process strengthens the gel, so that shrinkage during the drying step is kept to a minimum. (c) Drying of the gel . In this step, the gel should be freed of  the pore liquid. To prevent the collapse of the gel struc-

11

ture, drying is made to take place under special conditions. All methods of aerogel production involve these three general steps. Additional procedures can also be undertaken in order to influence the final product structure. There are, currently, a variety of methods for silica aerogel synthesis which will be reviewed and discussed in the following sections. 2.1.

Gel preparation

2.1.1. Starting materials Sol–gel precursors are mainly silicon alkoxides which can be obtained in a high degree of purity whereas potassium silicate is very difficult to purify. Tetramethoxysilane (TMOS) undergoes a more rapid hydrolysis than tetraethoxysilane (TEOS) and was extensively used by a Japanese team to produce monolithic silica aerogels (Nakanishi et al., 1998; Minakuchi et al., 1996; Nakanishi et al., 1997; Minakuchi et al., 1997; Ishizuka et al., 1998). Wagh et al. (1999) compared the aerogels obtained from three different precursors: TEOS, TMOS and PEDS (polyethoxydisiloxane) and claimed that TMOS yields narrow and uniform pores and higher surface area than TEOS. Methyltriethoxysilane (MTES) was used by Colon et al. (1996) f or the more flexible network produced from its and higher surface area than TEOS. A mixture of TMOS and MTES was also investigated by Harreld et al. (2002) and Ishizuka et al. (2000). Adding methyltrimethoxysilane (MTMS) to TMOS or dimethyldiethoxysilane to TEOS ( Cao and Zhu, 1999) increases the hydrophobicity of the aerogel and shifts the pore size distribution towards larger pore radii (Rao and Haranath, 1999). Rao et al. (2006) synthesized highly flexible and superhydrophobic silica aerogel just by MTMS precursor. They investigated the absorption and desorption of organic liquids in this type of aerogels. Their results showed that these aerogels absorbed the organic liquids and oils by nearly 15 times its own mass (Rao et al., 2007a). Functional groups on silicon atoms which impart the hydrophobic properties to the solids are atoms showing high electronegativity such as F. One of the sol–gel precursors may be a mixture of  alkoxysilanes, one of them being a functionalized moiety such as aminopropyltriethoxysilane (Mansur et al., 2000), (3(2-aminoethylamino) propyltrimethoxysilane (EDAS) (Alie et al., 1999) or N-octyltriethoxysilane as proposed by Rodriguez and Colon (1999) or Constantin and Freitag (2000) who performed a copolymerization process. Guo et al. (1999) used a mixture of TEOS and polydimethylsiloxane. Einarsrud et al. (2001) studied the hydrolysis of PEDS and Deng et al. (2000) prepared low cost silica aerogel with industrial PEDS (E-40) and found that, the thermal conductivity of the silica aerogels prepared with PEDS is a little higher than that prepared with TEOS. Recently, Zhou et al. (2007) synthesized hydrophobic aerogels from PEDS (E-40) and used perfluoroaklysilane (PFAS) as a coprecursor. The use of PFAS as a coprecursor with PEDS resulted in hydrophobic silica aerogels. They found that it had the largest surface area(1093.28m 2 /g)when the PFAS/E-40 volume ratio is at 0.6. Tang and Wang (2005) prepared the silica aerogel from rice hull ash that is one of the waste materials. Alkoxides are expensive and hazardous materials, which prohibit commercialization. These materials are not only

12

j o u r n a l o f m a t e r i a l s p r o c e s s i n g t e c h n o l o g y 1 9 9 ( 2 0 0 8 ) 10–26

As water and alkoxysilanes such as TEOS and TMOS are partially immiscible, additional solvent is needed to homogenize the mixture. A variety of solvents are used for this purpose, alcohols, acetone, dioxane, tetrahydrofurane being a few examples. In the case of alcohols, it should be noted that although they are used as solvent, they can participate in an esterification reaction and thus reduce the hydrolysis rate. The typical phase diagram of the system alkoxylane–water–solvent is shown in Fig. 1 (Brinker and Sherer, 1990). In their gel preparation, Zhang and Huang (2001) used methylene chloride to dissolve MTES. Dai et al. (2000) promoted ionic liquids. Einarsrud et al. (2001) dissolved the acid catalyst in ethyl acetoacetate and added this mixture to PEDS. Hydrolysis of silicon alkoxides is a versatile technique which can produce different materials according to different parameters and acid or base catalysis reaction. The critical point is Si:H2 O ratio. On the other hand, the proportions of  TMOS (or TEOS) and water is important and yields different products. Karmakar et al. (2000) observed that any acid–water mixture irrespective of the type of acid (formic acid, acetic acid, propanoic acid, pentanoic acid, hydrochloric acid, nitric acid, sulphuric acid, orthophosphoric acid) can produce silica

expensive but also hazardous and can cause blindness (with TMOS). To avoid the costs of raw materials, Schwertfeger et al. (1998) used waterglass as a cheaper silica source. Waterglass (or sodium silicate) solution was usedfor low-costaerogel production by Kistler in 1930s using super critical drying (SCD) method and now is widely used as the general material for commercial synthesis of silica aerogels by subcritical or ambient pressure drying (APD) method. Einarsrud and coworkers (2001) investigated properties of aerogels synthesized by different precursors and reported that waterglass, produced aerogels with the highest degree of monolithicity because these gels had the highest stiffness and the largest pore size. The Na2 O/SiO2 mole ratio, Na2 SiO3 /H2 O and silica content are important parameters for preparation of waterglass solution. Rao et al. (2004a) reported that, with Na2 SiO3 /H2 O molar ratio>8 × 10−3 , aerogels have optimum hydrophobicity and physical properties. Also Hwang et al. (2007) f ound that, the best value of Na2 O/SiO2 is 1:3.3 and the best silica content in solution is 4–8%. 2.1.2. Sol preparation and gelation In the case of alkoxysilane silicon derivatives, many procedures have been used by different investigators ( Table 1).

Table 1 – Sol-gel preparation

Silicon derivative TMOS 1a MTMS 1a PEDS 1b PEDS (E40) 0.2b PFAS/PEDS 0–1a MTES 0.75 a TMOS 1a TMOS 1a TMOS 1a TEOS 1a 1 1 1 1 TMOS 1a TEOS 1b TMOS 4b TEOS 1gb a b

Molar ratio. Volume ratio.

Solvent EtOH 4 MeOH 14–35 Ethyl acetate 1 EtOH 1 EtOH 8 CH2 Cl2 2 Ionic liquid 1 aceton 20 MeOH 0.5–3 EtOH 3 3 3 3 3 MeOH 4 EtOH 20 EtOH 4 EtOH 1 ml

Water

Catalyst

Additives

Reference

HCl 10−2 M

(Rao and Haranath, 1999)

Oxalic acid, NH4 OH

(Rao et al., 2006)

HF 21 N 0.02% HF

(Einarsrud et al., 2001)

HF 0.1 TFA

(Zhou et al., 2007)

1 8

(Deng et al., 2000)

3 0.5

(Zhang and Huang, 2001)

0.1 HCO2 H 2 HCO2 H

(Dai et al., 2000) (Moner-Girona et al., 2003)

2–8 2–20

CH3 CO2 H 0.06–1 M NH 4 OH 0.02–1 M

(Nicolaon and Teichner, 1968) (Kirkbir et al., 1996 )

4 4 4 4 4

HCl 0.02 HNO3 0.02 H 2 SO4 0.02 Oxalic acid 0.02 HF 0.02 NH4 OH 10−2 M

(Dieudonne et al., 2000)

4 NH4 OH(27%), HCl 1 M 2.8/3.2 HCl 10−2 M, NH4 OH 10−1 M

(Wu et al., 2000) (Mezza et al., 1999)

1 HCl 4.8 × 10−6 M 0.86 l

PEG

( Judenstein et al., 1994 )

  journal of materials processing technology 199 (2008) 10–26

13

Fig. 2 – Dependence of the relative hydrolysis and condensation rates on the pH of the solution.

Fig. 1 – Ternary phase diagram of the system TEOS–ethanol–water at 25 C. ◦

microspheres from TEOS with a water: TEOS ratio in the range 1.0–1.5 for strong acids and 1.5–4.0 for weak acids. MonerGirona et al. (2003) investigated the effect of water/TMOS ratio on the size of silica aerogel microparticles and found that higher ratios give smaller particles. Nicolaon and Teichner (1968) carriedout a study on the influences of waterand TMOS concentrationwhen producing aerogels exhibiting highporosity and concluded that the water quantity should be 2–5-time the stoichiometric proportion with a TMOS concentration in the alcohol of 5–10%. The molar ratio of H 2 O: Si(OR)4 in the sol should be at least 2:1 to approach complete hydrolysis of the alkoxide. By increasing water concentration, chemical reactions are accelerated and gelation times decreased. Hydrolysis is performed with a catalyst. Three procedures are proposed: acid catalysis, base catalysisand two-step catalysis. It is generally agreed that under acid catalysis, entangled linear or randomly branched chains are formed in silica sols whereas under base catalysis, it is easy to form a network of  uniform particles in the sol. Acid catalysis is performed with HCl, H2 SO4 , HNO3 , HF, oxalic, formicand aceticacids. A typical volume ratio of TEOS:C2 H5 OH:2 O:acid is 1:3:4:0.002. Gelation times aregenerallylonger when the pHof thesol is low. Kirkbir et al. (1996) observed that HF catalysis yields the highest pore volume and pore diameter but the gel is weak. They also promoted double acid catalysis (Murata et al., 1994) with HF and either HCl, HNO3 or H2 SO4 . Hydrolysis of TMOS at 70–80 ◦ C in an open vessel accelerates condensation and reduces the amount of liquid by expelling excess methanol through distillation (Bryans et al., 2000). Acid-catalyzed hydrolysis can be stimulated by sonication (Vollet et al., 2001). Base catalysis usually involves dilute ammonia 10−2 M (Dieudonne et al., 2000). Under base catalysis it is easy to form a network of uniform particles in the sol, and the resulting pore volume is quite large. In order to prepare hydrophobic aerogels Wagh and Ingale (2002) f ound that, theoptimal molar ratio of TMOS:MeOH:H2 O:NH4 OH and MTMS as a hydrophobic reagent was 1:12:4:3.6 × 10−3 :1.3, respectively.

Theinfluence of the pH on thegelationprocesswas studied by Brinker et al.(1984). They proposedtwo-step procedure that others used it (Wu et al., 2000; Mezza et al., 1999; Hafidi Alaoui et al., 1998; Rao et al., 2005a). In a typical procedure (Rao et al., 2005a) in the first step TEOS, ethanol, oxalic acid and H 2 O were mixed in themolarratio 1:8:6.23 × 10−5 :3.75, respectively. InthesecondstepamixtureofH 2 OandNH4 OHwiththemolar ratio of 2.25:4 × 10−2 was added to the silica sol (prepared in the first step). The addition of NH4 OHas a second catalyst to a sol initiallycatalyzed by HCl can increasethe rateof condensation reactions and reduce the gelation time. Gelation can take place on a wide time scale: seconds, minutes, hours, days, or months. Gelation times can be followed by measurement of  viscosity. The gel exhibits a Newtonian viscosity in the initial state and then transforms into a viscoelastic gel. Thegeneraldependenceof therelative rate of reaction (vrel ) for both hydrolysis and condensation is presented in Fig. 2.It is shown that under base catalysis condensation kinetics are faster than hydrolysis kinetics. Pope and Mackenzie (1986) has shown that thepH is notthe only factorcontrollingthe hydrolysis, gelation and thus, the properties of silica gels. One other factor is thenature of thecatalysts. A systematic investigation of 6 different catalysts used in the sol–gel process of TEOS in ethanol was carried out. It was found that the gelation time increasedin the following sequence for different acids used at the same concentration. HF < CH3 COOH < HCl < HNO3 < H2 SO4 The pH in turn, increases as follows: HCl, HNO3 , H2 SO4 < HF < CH3 COOH In general, acid-catalyzed hydrolysis and condensation lead to weakly branched and microporous structures, whereas basic conditions or two step acid–base processes increase cross-linking, leading to decreased microporosity and a broader distribution of larger pores in silica gels (Rao et al., 2004a; Rao et al., 1999; Rao and Bhagat, 2004 ). In the case of waterglass based aerogels, the sol preparation is quite different. The sol in this case is silicic acid (H2 SiO3 ) which is produced by exchanging Na + ions of waterglass with H+ . To remove sodium ions in the waterglass, the dilute sodium silicate is passed through ion-exchange col-

14

j o u r n a l o f m a t e r i a l s p r o c e s s i n g t e c h n o l o g y 1 9 9 ( 2 0 0 8 ) 10–26

Fig. 4 – Relative aging rate as a function of time for two aging mechanisms: (a) Reprecipitation of silica dissolved from the particle surfaces onto the necks between particles. (b) Reprecipitation of small dissolved silica particles onto larger ones. Fig. 3 – TEM image of silica sol during gelation.

umn filled with strong acidic cation resin such as sulphonated polystyrene type. The pH of waterglass solution is ∼11.5 and the pH of sol after ion-exchanging is ∼2.5 (Lee et al., 2002). After sol preparation, catalyst is added for faster gelation. The usual base catalyst is ammonium hydroxide (NH 4 OH). The TEM micrograph of silica sol is shown in Fig. 3.it can be seen that the colloidal silica particles size is 70–80 nm in diameter. During the gelation and aging procedures, syneresis of wet gels was observed and the linear shrinkage of wet gel caused by syneresis was determined by Hwanget al. (2007)to be about 10%. According to references (Lee et al., 2002; Judenstein et al., 1994), the optimum pH, gelation temperature, and aging time were 3.5, 60 ◦ C, and 24 h, respectively. Lee et al. (2002) f ound thatporosities,densities, and surface areas tended to decrease with the increase of sol pH. 2.2.

Additives

The pore size and the mechanical properties of gels can be varied with the addition of polyethylene glycol (PEG) to the sol. PEG is a porogen which acts as a through-pore template and solubilizer of the silane reagent. This has been done by Nakanishi et al. (1998), Judenstein et al. (1994) and Martin et al. (2001) who claimed that high concentrations of PEG weaken the solid matrix whereas small concentration of PEG strengthen the matrix. The pore size of macroporous silica aerogel can be controlled by varying the concentration of  water soluble polymer. Reetz et al. (1996) used some additives as polyvinyl alcohol (PVA) and PEG as biocatalyst supports, because it was found that the lipase activity was significantly enhanced in the presence of these additives. Narrow and more uniform pore size distribution was observed with the addition of glycerol which acts as a drying additive since it prevents further reaction of water (Rao and Kulkarni, 2003). Use of a surfactant in water–ethanol solution produces surfactant templated aerogels ( Anderson et al., 1998). Since the hydrophobic properties of TMOS based aerogels are increased by incorporating MTES as

a synthesis component (Rao and Kulkarni, 2002), Alie et al. (Alie et al., 2001a,b; Alie et al., 2002) considered the incorporation of other silanes to be equivalent to the incorporation of additives. Differences in reactivity are obvious. Consequently, the nucleation mechanism is related to the difference in reactivity between the main reagent (TMOS, TEOS, or TPOS) and the additive. They examined the behavior of 3-(2-aminoethylamino)propyltrimethoxysilane(EDAS) (Alie et al., 1999), 3-aminopropyltriethoxysilane (AES), 3aminopropyltrimethoxysilane (AMS), propyltrimethoxysilane (PMS) and 3(2-aminoethylamino) propyltriethoxysilane (EDAES). The main reagent is TMOS and TEOS. A nucleation mechanism by the additive takes place. When the additive contains methoxy groups, it reacts first to form particles on which the main reagent condensesin a later stage. When both the additive and the main reagent contain an ethoxy group, there is no nucleation mechanism by the additive. Amine or alkyl groups only influence gelation time. Addition of  silica spheres (aerosil) in the solution before gelation strongly affects the aggregation mechanism; two fractal structures coexist. 2.3.

Aging

Two different mechanisms might operate during aging that affect the structure and properties of a gel; (a) Neck growth from reprecipitation of silica dissolved from particle surface onto necks between particles. (b) Dissolution of smaller particles and precipitation onto larger ones. These two mechanisms will operate simultaneously, but at different rate, as illustrated in Fig. 4 (Strøm et al., 2007). Washing in H2 O/EtOH increases the liquid permeability of the solid part of the gel by a dissolution reprecipitation process for silica. Aging in a siloxane solution increases the stiffness andstrengthof thealcogel by addingnew monomers to the silica network and by improving the degree of siloxane cross linking; conversely this stepwill reduce the permeability (Kirkbir et al., 1998). During aging, material is transported to the neck region between particles giving a more rigid gel network. The driving force for thematerial transport is difference in solubility, S, for surfaces with different curvatures, r, given

  journal of materials processing technology 199 (2008) 10–26

by the Kelvin Eq. (1) (Brinker and Sherer, 1990). S = S0 exp



2 sl Vm RTr



(1)

where S0 is the solubility of a flat surface of the solid phase,  sl is the solid–liquid interfacial energy, Vm is themolar volumeof  the solid, R is the ideal gas constant, and T is the temperature. Necks between particles have a negative curvature ( r <0) and hence a low solubility. Material will accumulate in these convex areas after being transported from the concave surface of a particle. The smaller particles have larger solubility. So, the driving force will also act to dissolve the smallest particles followed by precipitation onto larger particles. This ripening mechanism will, however, lead to a coarsening of the structure and is the result after too long aging time. ( Hæreid et al., 1995). Titulaer et al. studied the fluid composition effects on silica gel aging by surface area analysis (Titulaer et al., 1994). Solvent effect on aging of silica gels was studied by Chou and Lee (1994). They related the surface area to the polarity parameter of the aging solvent. Effect of aging and pH on the modulus of aerogels was studied by Hdach et al. (1990). The strength and stiffness of the wet gels as a function of  aging time, temperature and pH in the TEOS/ethanol solution were also described(Haereid et al., 1996, 1995a). Einarsrud and Nilsen (1998), Einarsrud (1998) and Einarsrud et al. (1998) in a detailed study showed that, gels synthesized by waterglass and colloidal sol can be strengthened by aging in TEOS, water and ethanol solutions. Gels were washed with a 20% water–ethanol solution for24 h at 60 ◦ C, then an aging solution (70%TEOS:ethanol, v/v) was used for 6–72 h at 70 ◦ C followed by washing with ethanol and heptane. Data from small-angle neutron scattering showedonly a slight increasein the volume fractal dimension of the porous gel network. The same group demonstrated that washing in a water solution increases the permeability of the gels by dissolution reprecipitation (Ostwald ripening) (Einarsrud et al., 2001). Effect of concentration of aging solution and aging time on the porosity characteristics of silica aerogels was investigated by Smitha et al.(2006). They found that bulk density andlinear shrinkage decreased whereas surface area, pore size and pore volume increased with increase in concentration of TEOS in the aging solution. Aging time also has a similar effect on bulk density, surface area and pore volume. The best concentration of TEOS in aging solution and time of aging were reported as 80% and 48 h, respectively. Einarsrud et al. (2001) observed that stiffness of the gels prepared by PEDS can be increased by soaking in the PEDS solution. They found that such aging increases both cluster and particle dimensions. Low-density aerogels with a surface area of  ∼700m2 /g have been reported. Strøm et al. (2007) optimized this work by decreasing the concentration of PEDS in the aging solution,which avoids creation of a pore size distribution gradient between the surface and bulk of the gel. Influence of the aging process on the microstructure of  NH4 OH- and NH 4 F-catalyzed gels was studied by Suh et al. (1999). It was shown that the aging procedure allows the reorganization of the structure, resulting in a unimodal pore size distribution. Thepore volumeand mean pore diametersof the

15

resulting aerogels increase, whereas the surface areas remain virtually unaffected. During theaging process, the solvent evaporates, causing a slightshrinkage of thenetwork beforethe completion of aging process. Thus, if solvents with a low vapor pressure are used, this kind of shrinkage can be eliminated. Ionic liquids were used for this purpose by Dai et al. (2000). Ionic liquids are a class of solvents that have extremely low vapor pressure and possess versatile solvent properties. As a consequence, they offer an attractive method for achieving longer gelation times without shrinking of the gel network. The gels synthesized by this method are so stable, that even conventional drying causes just a little shrinking. This method can be helpful for the preparationof aerogels and aerogel film having a moderate density. Reichenauer (2004) showed that aging of silica gels in water not only increased mechanical stability, but also led to a decrease in external and micropore surface area which allows supercritical drying without significant shrinkage of the gel. Hæreid et al. (1995) investigated the effect of aging alcogels in water and found a maximum in both G and MOR independent of the aging temperature, but the aging time required to reach the maximum decreased with increasing temperature. Recently, Estella et al. (2007) studied the effect of some aging medias such as EtOH and NH 3 (aq) on the structural and textural properties of silica aerogels and found that increasing of micropore volume for alcogels that aged in ammonia is higher than those aged in EtOH and samples aged in NH 3 (aq) had a larger density than those aged in ethanol. Lately Strøm et al. (2007) studied the three different aging routes: (1) aging in sealed mould; (2) aging in solvent; and (3) aging in simulated pore liquid. They found that, all aging processes gave stronger and stiffer wet gels. However, a maximum in strength andstiffnesswas observed after a certain aging time.The third route gave a shortening in aging time with a maximum in MOR and G modulus after 8 h or less. Rao et al. (2004a) investigated the influence of gel aging on properties of waterglass based aerogels and found that increasing theperiod of gel aging resulted in increasedshrinkage with an increase in density and the corresponding optical transmittance increased. In order to obtain transparent silica aerogels of low density and larger size without cracks, the gel aging period must be long enough so that the modulus of the wet gel increases thereby leading to monolithic aerogels. It should be mentionedthat in many references the aging procedure is not mentioned.

2.4.

Drying

Drying of the gel is a critical step. Drying is governed by capillary pressure. Shrinkage of the gels during drying is driven by the capillary pressure, Pc , which can be represented by (2).

Pc

=

− lv

( r p − ı)

(2)

where  lv is the surface tension of the pore liquid, rp is the pore radius,which can be represented by (3). ı is the thickness of a surface adsorbed layer (Brinker and Sherer, 1990; Scherer

16

j o u r n a l o f m a t e r i a l s p r o c e s s i n g t e c h n o l o g y 1 9 9 ( 2 0 0 8 ) 10–26

et al., 1996). rp

=

2Vp Sp

(3)

where Vp and Sp are pore volume and surface area, respectively. They are critical parameters. It is the gradient in capillary pressure within the pores that leads to mechanical damage; the capillary tension developed during the drying may reach 100–200 MPa (Scherer and Smith, 1995) with consequent shrinkage and cracking. The small pore size can induce fracture during drying due to enormous capillary forces. In supercritical drying (SCD) method the pore liquid is removed above the critical temperature ( Tcr ) and critical pressure (Pcr ) of the concerned liquid. At this point there is no liquid–vapour interface and, thus, no capillary pressure (Brinker and Sherer, 1990). By drying at ambient pressure, the surface tension between liquid and vapor cannot be avoided. Stress within the gel is proportional to the viscosity of the pore liquid, the drying rate and inversely proportional to the permeability of the wet gel. The important parameters are: the pristine gel strength; the pore size of the wet gel; and the solvent used in drying. The small pore size can induce fracture during drying due to enormous capillary forces. The pore liquid is under enormous ˚ On the other tension when the pore size is smaller than 200 A. ˚ the shrinkage hand, when the pore size is larger than 200 A, will be less and cracking will be less likely to occur ( Rao and Haranath, 1999). Conversely in some cases, small pore size ˚ are easier to dry than larger pore size gels which is gels (40 A) explained by a theory of cavitation (Scherer and Smith, 1995). Manipulation of pore size distribution can be done through the drying solvent. 2.4.1. Supercritical drying (SCD) There are two different methods of supercritical drying: high temperature (HTSCD) and low temperature (LTSCD). 2.4.1.1. High temperature supercritical drying (HTSCD). HTSCD method is the first method of drying the silica aerogel. This method was applied by Kistler in 1931 and is still widely used for silica aerogel production. HTSCD is schematically represented in Fig. 5. The process is carried out in three steps as follows:

ture is slowly raised. This causes a pressure increase. Both the temperature andpressure areadjustedto reach values above the critical points of the corresponding solvent. On attainingthe set temperature and pressure, the conditions are kept constant for a certain period of time. (2) The fluid is then slowly vented at constant temperature, resulting in a pressure drop. (3) When ambient pressure is reached, the vessel is cooled to room temperature. Thus, the phase boundary between liquid and gas is not crossed during the drying process. It is important to note that supercritical drying in organic liquids leads to rearrangement reactions in the gel network due to high temperature conditions. For example, under supercritical conditions, the silica aerogel surface is reesterified, making the material hydrophobic and stable when exposed to atmospheric moisture. Gross et al. (1998) conducted a “rapid supercritical extraction process” where the rates of condensation reactions were increased due to the increased temperature. Sol was poured directly into a container and heated immediately to supercritical conditions in an autoclave. Gelation and aging occurred during heating and the reaction rates were found to be very high due to the high temperatures. Also, the gel filled the container completely, which enabled relatively fast venting of the supercritical fluid. This process seems to be an interesting and practical modification of normal high temperature drying. This method, however, may present problems due to the combination of  high temperatures and high pressures as well as the flammability of the solvents. Many efforts have been undertaken for establishing a drying procedure that can be carried out at moderate temperatures and pressures. Kirkbir et al. (1998) observed that a threshold pressure exists above which the shrinkage is negligible. Below thethreshold point, the capillary pressure overcomes the strength and the structure collapses. The threshold point depends on sol composition. They managed to reduce the drying pressure (but not the temperature). Several different solvents such as ethanol, butanol, pentanol and isooctane were tested and, in each case, it was found that the drying pressure could be reduced to a certain value, so that shrinkage of the aerogels did not exceed 5%. They found that shrinkage is negligible for wet gels dried in 2-pentanol at 1.8 MPa and 300 ◦ C. This can be explained by silica solubility (Table 2) and structural modification since the process may

(1) The wet gel, together with a sufficient amount of solvent (e.g. methanol) is placed in an autoclave and the temperaTable 2 – Silica solubility in n-alkanols Methanol (mg/l) Ethanol(mg/l) Propanol (mg/l)

Pore liquid

Ethanol Isobutanol Isooctane 2-Pentanol

Fig. 5 – Schematic procedure of high temperature supercritical drying.

1890 164 8

20 ◦ C ( lvdyne/cm)

22.4 22.6 24.0 18.8

205 ◦ C Pv

 lv

Pcr

3.2 1.8 1.0 1.0

2.5 7.2 6.1 3.8

0.7 1.9 1.6 1.0

 lv , surface tension, Pcr , capillary pressure, Pv , vapor pressure.

  journal of materials processing technology 199 (2008) 10–26

induce an increasein the network connectivityby formationof  new siloxane bonds. The variation in silica solubility may not be sufficient to explain the variation in shrinkage with type of  alcohols. Rapid supercritical extraction (RSCE) is another method for SCD. It wasdeveloped by Poco et al. (1996) and used by Scherer et al.(2002) and Gauthier et al.(2004). In this method, theliquid precursor chemicals and catalyst are inserted into a two-piece mold that is then heated rapidly to speed up the polymerization. The pressure is initially set by fastening the two mold partstogetherwith properlytensioned bolts,or byapplying an external hydrostatic pressure inside a larger pressure vessel, or by a combination of these two. Once the supercritical point of the alcohol is reached, the supercritical fluid is allowed to escape through gaps formed by the roughness in the surface contact between the two portions of the mold, or through a relief valve set just above the supercritical pressure. A benefit of this method is that the entire process is carried out in one step, and can be accomplished under an hour, as opposed to multiple steps for all other available methods. In addition to the reduction in fabrication time, the rapid supercritical extraction as a one-step process has the most potential for reliable and repeatable fabrication, as well as increased production volume. 2.4.1.2. Low temperature supercritical drying (LTSCD). An alternative method for aerogel drying was suggested in 1985 by Tewari et al. (1985). The solvent present in the gel before drying (generally alcohol) is replaced by a liquid having a critical point close to ambient temperature. Liquid CO2 was found to be the most practical choice. LTSCD has the advantage of being implemented at a low temperature (<40 ◦ C) and moderate pressure (<80bar). The whole process is shown schematically in Fig. 6. One of the experimen-

17

Fig. 6 – Schematic procedure of low temperature supercritical drying.

tal devices used for drying is shown in Fig. 7 (Wagh et al., 1999). The experimental procedure is as follows: (1) The gel containing excess amount of solvent (e.g. CH3 OH) is placed in an autoclave .The vessel is sealed and liquid CO2 is pumped in at 4–10 ◦ C until the pressure reaches about 100 bar. The outlet valve is then opened so that the solvent extracted by the liquid CO 2 , is able to flow out. When thesolventis completely replacedby CO2 , the pump is turned off; the temperature is raised to 40 ◦ C (i.e. above the critical temperatureof CO2 , Tc = 31 ◦ C) andthe pressure is kept constant at 100 bar. (2) On reaching 40 ◦ C, and thus ensuring the transition of CO 2 into the supercritical state, the system is slowly depressurized under natural flow. (3) When ambient pressure is reached, the system is cooled down to the room temperature. Aerogels obtained by this method are hydrophilic.

Fig. 7 – Schematic representation of the supercritical carbon dioxide equipment used for LTSCD.

18

j o u r n a l o f m a t e r i a l s p r o c e s s i n g t e c h n o l o g y 1 9 9 ( 2 0 0 8 ) 10–26

Ten years later, the LTSCD process was modified such as to involve the use of supercritical CO 2 as compared to liquid CO2 (van Bommel and de Haan, 1995). In this case, the heating andcooling steps canbe eliminated. A continuous process of aerogel production including the recycling of CO 2 was also suggested. The LTSCD process also involvesan extraction step, which strongly depends on the diffusion of CO 2 in the solvent. The two main transport mechanisms of alcohol and CO 2 are Knudsen and surface diffusion ( van Bommel and de Haan, 1995). It is difficult to predict the duration of the drying step, because in most cases the diffusion coefficients of the liquid in the sample are unknown. If the duration of diffusion is not longenough,a non-transparent areainside the gel is observed, or in theworst cases,cracking of theaerogel occurs. In orderto describe the extraction process, binary diffusion coefficients of methanol–liquid CO2 and methanol–supercritical CO2 were determined by Novak et al. (1999). They found that the diffusion coefficient increases with increasing temperature. But in low-temperature aerogel production, replacement of an original solvent with liquid or supercritical carbon dioxide is the critical stage andit is one fully controlled by diffusional mechanism. Wawrzyniak et al. (2001) investigated the variation of  the effective diffusion coefficient of ethanol in the vicinity of  the critical point of CO 2 . Kocon et al. (1998) f ound that shrinkage during supercritical drying is mainly due to the restructuring of clusters which strongly depends on the strength of bonds. Silica dissolution leads to the weakening of these bonds. Solvent composition and solvent quantity have been investigated to keep the silica dissolution as low as possible. Dieudonne et al. (2000) claimed that thermal treatment of silica gels under alcohol in an autoclave induces textural transformations of the solidnetwork on a nanoscopic scale. Smith et al. (1995) developed a model to predict gel shrinkage from which it turns out that the number of process variables which canbe utilized to control shrinkage is rather limited (pore fluid, drying rate, and initial density). Loy et al.(1997) tried to solve theshrinkage problem bysynthesizing silica aerogels directly in supercritical CO 2 , making solvent exchange unnecessary. In this case, the standard reactions cannot be applied, because water is produced and water is poorly miscible with CO2 under normal conditions. This problem wasavoided byapplying the “water-free” sol–gel polymerization technique, described by Sharp (1993), who applied the technique for aerogel production with supercritical drying. Loy et al. (1997) reported that TMOS reacts with formic acid in the presence of CO 2 as a solvent. After 8–18h of aging, the pressure was slowly released and white, opaque silica aerogels were obtained that could be explained by the relatively high acid concentration. Direct formation of silica aerogel powders in supercritical CO2 and acetone was reported by Moner-Girona et al. (2000). Unfortunately, until now, detailed information on the properties of the resulting aerogels have not been reported. 2.4.1.3. Comparison of HTSCD and LTSCD. A comparative study of resulted aerogels dried by HTSCD and LTSCD methods was performed by Ehrburger-Dolle et al. (1995). It was shown that the microporosity of CO 2 -dried aerogels is significantly larger than that of the corresponding MeOH-dried aerogels. However, the micro and mesoporous textures of the

CO2 -driedaerogel areequivalent to those of thealcogel. These results are in agreement with those of other authors ( Yoda and Ohshima, 1999). Shrinkage of the CO2 -dried aerogel is probably due to the reorganization of aggregates during the exchange of alcohol with liquid CO 2 . The aggregate arrangement is very similar to that observed in gels formed with pyrogenic silicas. In contrast, after HTSCD, the structure collapses at a short length scale and thus, the initial aggregate network is strengthened. Dieudonne et al. (2000) illustrated by small-angle X-ray scattering (SAXS) experiments, that the aerogels obtained by methanol supercritical drying show a smooth surface, whereas the CO 2 -dried aerogels have tough solid particles. However, transparency of the CO 2 -dried samples is comparable with that of alcohol-dried samples ( Tewari et al., 1985). This agrees with the results of  Tajiri and Igarashi (1998) who also studied thetransmittance of aerogels, dried in different supercritical media. It wasreported, thatfor thehightemperature SCD, isopropanol seems to be the most favored medium. A model relating thedrying stress to thethermalexpansion and flow of the pore liquid has been designed for comparison of the two drying techniques (Unsulu et al., 2001). This model suggests that the CO 2 exchange method results in smaller stresses andthat these stresses are of such a small magnitude that crack-free gels result. 2.4.2. Ambient pressure drying (APD) In spite of the differences between the HTSCD and LTSCD processes, both are considerably expensive due to the high pressures involved. For this reason there is a great interest in subcritical or ambient pressure drying (APD). These approaches offer great promise to lower costs for aerogel production and thusrepresent an important considerationfor the future development of these materials. Combining the possibilities for gel formation and drying leads to at least two possible routesfor ambient pressure dried silica aerogels (Fig. 8): 1. Alkoxysilane based aerogels. (Rao et al., 2005a; Davis et al., 1992a,b; Deshpande et al., 1992) 2. Waterglass based aerogels: this routewas used by Schwertfeger, Tyler, Rao and others. (Schwertfeger et al., 1998; Tyler, 1959; Tyler, 1962; Rao et al., 2004a). Ambient-pressure methods for silica aerogels include both surface modification and network strengthening. Additionally, the contact angle between the pore liquid and the pore walls has to be influenced so as to minimize capillary forces. This involves chemical modification of the inner surface, e.g., via silylation. The water–alcohol mixture in the pores of the gel is first exchanged for a water-free solvent. Then, reaction with silylating agent (such as: TMCS, HMDS, HMDZ . . .) takes place so that Si–OH groups are silylated. Silylation is carried out directly in the water phase of the hydrogel, which results in a solvent exchange as well as a phase separation of the gel water and the solvent. The replacement of H from the Si–OH groups by the hydrolytically stable Si–R groups through the oxygen bond prevents the adsorption of water and hence results in the hydrophobic aerogels (Fig. 9) (Schuth et al., 2002). The dipole

  journal of materials processing technology 199 (2008) 10–26

19

Fig. 8 – General routes for drying of silica aerogels.

moment of the hydroxyl group is higher than that of alkyl (–R) or alkoxy (–OR) group, and therefore it is expected that the dielectric constant of the SiO 2 aerogel decreases after silylation. The aerogels should have lower dielectric constant for higher hydrophobicity in them. After another solvent exchange, drying takesplace by evaporation. Evaporative drying of silica gels can be separated into three step: (Brinker and Sherer, 1990). (1) Constant rate period when the volume losses of the gel and evaporated liquid are balanced; (2) First falling rate periodwhen liquidflowsthroughpartially empty pores; (3) Second falling rate period when liquid escapes the drying gel only via diffusive vapor transport to the exterior surface. The last step is the most complex phase because several phenomena occur simultaneously. One of the important parameters in this step is the apparent diffusion coefficient of the solvent. Bisson et al. (2004) studied this phase of  evaporating for treated (or silylated) silica gels. They used thrimethylcholorosilane (TMCS) and isopropyl alcohol (IPA) as silylation agent and solvent, respectively. The diffusivity of solvent vapors was found to be very low for treated and

Fig. 9 – Silylating of silica gel.

untreated samples. They used microcalorimetry to confirm the reduction in the affinity between the solid phase and solvent due to the silanization treatment. The substitution of  silanol groups by trimethylsilyl groups diminishes the interaction strength between the solid phase and the solvent. Consequences of the surface modification are a reduction of  the maximum densification observed during drying, and an increase in the apparent diffusion coefficient. Comparison between untreated and treated samples during the last drying step showed how and when spring-back takes place for the latter (Bisson et al., 2004; Schadl and Mersmann, 1985). Deshpande et al. (1992) described a drying method based on the so-called “spring-back effect”. During drying the silylmodified gels begin to shrink due to the development of  capillary forces. However, a “spring-back” effect is observed when the liquid phase starts to form isolated droplets within the gel network. Since neighboring surface silyl groups are chemically inert and detach with little activation energy, the gel body is able to re-expand (Fig. 10). Rao et al. (2005a) prepared the two step processed ambient pressure dried silica aerogel withimproved properties in terms of low density, high hydrophobicity, low thermal conductivity, high porosity and high optical transmission at ambient pressure with TEOS precursor and hexamethyldisilazane (HMDZ) silylating agent. The best quality silica aerogels have been obtained with the molar ratio of TEOS:HMDZ at 1:0.36. Rao et al. (2001) studied the effect of silylating agents on properties of silica aerogels prepared by APD method and reported that, by using the TMOS precursor, TMCS resulted the best quality silica aerogels in term of monolithicity, visual transparency and lowest density. The APD process relies on a modification, usually silation of the internal gel surface and drying at ambient pressure. Here, the gel still shrinks during drying. In order to prevent the condensation of reactive groups on walls of shrunk pores

20

j o u r n a l o f m a t e r i a l s p r o c e s s i n g t e c h n o l o g y 1 9 9 ( 2 0 0 8 ) 10–26

Fig. 10 – Spring-back phenomena.

the surface modification must provide surface passivation. Because silylation requires an organic solvent, surface modification occurs in an organic solvent and solvent exchange becomes necessary. The solvent exchange was applied by multi-step process. In order to solve problems of multi-step solvent exchange needing a very long diffusion time and consuming too many solvents, Schwertfeger et al. (1998) developed a process in which one-step solvent exchange and surface modification were simultaneously progressed using HMDSO (hexamethyldisiloxane)/TMCS solutionfor modification of the wetgels using waterglass as a cheap silica source. These silylation agents not only achieve the desired surface modification, but also react with the water in the pores of the hydrogel to form a low tension organic solvent, suitable for ambient drying conditions. Kim and Hyun (2003) and Lee et al. (2002) also synthesized silica aerogels via one-step solvent exchange/surface modification of wet gels using IPA/TMCS/ n-hexane solution. Hwang et al. (2007) f ound that the best TMCS/pore water is between 0.3 and 0.4. Shi et al. (2006) repeated this work using EtOH/TMCS/heptane solution. The modification mechanism by this type of solutions can be illustrated as in Fig. 11.These aerogels are normally hydrophobic. The Hoechst group has also shown that these aerogels can be turned into hydrophilic materials by oxidation of the surface groups (Schwertfeger et al., 1996). Rao et al. (2007b) tested the thermal stability of the hydrophobic aerogels at various temperatures and found that aerogels retained the hydrophobicity up to a temperature of 325 ◦ C in air and to 450 ◦ C in N2

atmosphere and lost it above this temperature. This is due to the fact that at this temperature, the Si–CH 3 groups get oxidized into Si–OH, which are responsible for adsorption of  water. Recently, Rao et al. (2004b) investigated the properties of ambient pressure dried silica aerogels with waterglass precursor using various silylating agents .The physical properties of the aerogels such as density, porosity, pore volume, thermal conductivity and contact angle measurements were studied by using various mono, di and tri alkyl or aryl silylating agents such as: vinyltrimethoxysilane (VTMS), phenyltrimethoxysilane (PTMS), phenyltriethoxysilane (PTES), dimethyldimethoxysilane (DMDMS), trimethylmethoxysilane (TMMS), bis(trimethylsilyl)acetamide (BTSA), MTMS, MTES, TMCS and HMDZ. They found that, the best quality was obtained with tri alkyl silylating agent such as HMDZ agent. Rao et al. (2001) reported the same results. They also reported that using n-hexane or n-heptane as the solvent resulted in aerogel with the lowest density and best visual transparency. In general two different methods are used for the surface modification of the gels. (1) Co-precursor method: in this method, the surface-modifying agent is added to the silica sol itself before gelation. (2) Surface derivatization method: according to previous discussions, in this method, the gel is obtained first and then kept in a bath containing a mixture of solvent and surface modifying agent. The latter has been used extensively for the synthesis of the water–glass based aerogels via APD (Rao et al., 2005a; Haereid et al., 1995b;

Fig. 11 – Schematic presentation of the reactions occurring during the modification process.

21

  journal of materials processing technology 199 (2008) 10–26

Rao et al., 2005b). In this method, the mass transfer in the gel takes place by diffusion only. Large amounts of solvent and significant time are required to achieve solvent exchange and subsequent surface modification, respectively and it is very costly. The co-precursor method offers an advantage of  rapid surface modification in the gels for which the surface derivatization method takes long time. Moreover, the addition of the surface modifying agent to the sol (i.e., co-precursor method) results in the surface as well as the bulk modification uniformly throughout the gel. By this method, Bhagat et al. (2006,2007) was able to reduce the processing time (down to one day) with the employment of co-precursor method for surface modification in hydrogels using TMCS and HMDZ. Densities for ambient pressure dried gels are as low as 0.150 g/cm3 . It was shown that the total pore volume obtained by this method is even larger than that of CO 2 -dried samples of the same composition (Land et al., 2001). 2.4.3. Freeze drying Another drying method where the phase boundary between the liquid and gas phase does not exist and thus the capillary pressure does notplay an importantrole,is freeze-drying. Here, the solvent must be exchanged with a low expansion coefficient and a high sublimation pressure. The pore liquid is frozen and sublimed under vacuum ( Husing, 1997). Material obtained by this way is called a cryogel. Unfortunately, freeze drying has many disadvantages including the fact that the aging period has to be prolonged for stabilization of the network and, in some cases; the network may be destroyed by crystallization of the solvent in the pores.

3. Properties of silica aerogels and methods of determination Extensive interest in aerogels, particularly silica aerogels, is due to their unusual solid material properties. In this section, the properties of silica aerogels and the methods used to determine such properties, are presented. Table 3. provides an overview of the most important physical properties of silica aerogels. 3.1.

Pore structure

According to IUPAC classification for porous materials, the pores less than 2nm in diameter are termed “micropores”; those with diameters between 2 and 50nm aretermed ”meso-

pores“, and those greater than 50 nm in diameter are termed ”macropores“. Silica aerogels possess pores of all three sizes. However, the majority of the pores fall in the mesopore range, with relatively few micropores (Husing and Schubert, 1998). The approximate values of the pore size are between 5 and 100 nm, with an average pore diameter between 20 and 40 nm and a BET surface area between 600 and 1000m 2 . Porosity can be as high as 99%. One important aspect of the aerogel pore network is its ”open“nature and interconnectivity. In open-pore structures, however,fluids can flowfrom poreto pore, withlimited restriction, and eventually travel through the entire material. These pore structures of silica aerogels, lead to applications as filters (Cooper, 1989), absorbing media for desiccation (Gesser and Goswami, 1989; Liu and Komarneni, 1995; Komarneni et al., 1993) and waste containment (Attia et al., 1994). Aerogels have been recognized for many years as excellent catalysts and catalyst supports ( Schneider and Baiker, 1995; Dunn et al., 2004). It is of extreme importance to state the method of determination used when stating porosity data. Aerogels have an unusual combination of high porosity and small pore size, making porosity characterization by conventional techniques such as mercury intrusion (MIP), thermoporometry (TPM), and nitrogenadsorption/desorption (NAD), difficult. All these techniques are based on the application of capillary pressures on the aerogel network, which may cause large volumetric compressions, leading to incorrectvalues for pore size andvolume (Scherer et al., 1995). The mostwidely utilizedmethod for determinationof aerogel porosity is the nitrogen adsorption/desorption technique or BET method. (Shi et al., 2006; Rao and Wagh, 1998; Kim and Hyun, 2004). In this method, the amount adsorbed gas is measured (Richard Brundle and Evans, 1992). Nitrogen-adsorption techniques can, in principle, coarsely distinguish various pore shapes by the shape of the isotherms ( Sing et al., 1985). However, apart from experimental deficiencies ( Reichenauer and Scherer, 2001), the shapes of aerogel isotherms are very similar. High resolution scanning electron microscopy (SEM) and transmission electron microscopy (TEM) are techniques that may provide information on the microstructure of aerogels. These methods also present difficulties, mainly related to sample preparation. During the experiment, the fine aerogel powder may become electrostatically loaded and thus altered, in some way. These methods, however, allow a direct examination of the aerogel structure allowing the particle size and

Table 3 – Typical properties of silica aerogel

Property Apparent density Internal surface area %Solids Mean pore diameter Primary particle diameter Refractive index Coefficient of thermal expansion Dielectric Constant Sound velocity

Value 0.003–0.35 g/cm 3 600–1000 m 2 /g 0.13–15% ∼20 nm 2–5 nm 1.0–1.08 2.0–4.0 × 10−6 ∼1.1 100 m/s

Comments Most common density is

0.1g/cm 3

∼

Typically 5% (95% free space) As determined by BET method Determined by electron microscopy Determined using ultrasonic methods For a density of 0.1 g/cm 3 For a density of 0.07 g/cm 3

22

j o u r n a l o f m a t e r i a l s p r o c e s s i n g t e c h n o l o g y 1 9 9 ( 2 0 0 8 ) 10–26

Fig. 12 – (a) SEM; (b) TEM picture, showing the pore characteristics of silica aerogels.

pore size to be approximately estimated. TypicalSEM andTEM pictures of a silica aerogel are shown in Fig. 12 (Moner-Girona et al., 2003). The deduction of information about a three-dimensional object froma two-dimensional image is also speculative, especially for highly porous materials. The term “pore size” in aerogels is even more difficult to comprehend, considering that theporous networkhas a fractal nature (Gimel et al., 1995; Stauffer and Aharony, 1994) which confers scale-invariant properties over a large length scale (arguably fromnanometers to micrometers). Positron annihilation spectroscopy (PAS) is a nondestructive technique for the characterization of open volume defects in materials(from single vacancies to mesopores) (Schultz and Lynn, 1988; Puska and Nieminen, 1994). The unique interactions of the antimatter probe – the positron – with material electrons give PAS extreme sensitivity limits with respect to small defect size at low density, surpassing that of other techniques (PositronMaterials, in press). The positron lifetime, number of annihilation photons, and electron momentum are measured by different experimental techniques, and used to deduce different porosity characteristics ( Schultz and Lynn, 1988; Puska and Nieminen, 1994). Lately Mincov et al. (2004) utilized positron annihilation lifetime spectroscopy (PALS) for non-destructive characterization of porosity in aerogels. Small-angle light scattering has been found to be a good solution to the problem of pore structure characterization in aerogels. This technique is not destructive, and neither sensitive to pore connectivity. By simple alteration of the instrumental set-up and wavelength, one is able to sample a considerably large size range (Hua et al., 1995; Hunt, 1998). Also the structure of the SiO 2 -network can be investigated by making SAXS and ultra small-angle X-ray scattering (USAXS) (Dieudonne et al., 2000; Barbieri et al., 2001; Platzer and Bergkvist, 1993; Posselt, 1991). USAXS and atomic force microscopy (AFM) can be used to check densification. 29 Si MAS nuclear magnetic resonance (NMR) canbe used to identify andcharacterize theearlystages of condensation reactions. These methods (SAXS, NMR) of  investigation are difficult to apply to the study of large monolithic aerogels (Emmerling et al., 1993; Glatter and Kratky, 1982).

3.2.

Density

Volume shrinkage of the aerogels is calculated from the volumes of the hydrogel and aerogel. Two different terms are used to characterize silica aerogels: bulk density and skeletal density. Bulk density (b ) is definedas theratio of theaerogel’s mass to its volume. The texture of the solid part of aerogels is made of ultra fine particles. The skeletal density of these particles is supposed to be very close to that of the bulk solid. These values were obtained by using helium picnometry (Woignier and Phalippou, 1987). The percentof volume shrinkage, porevolume andporosity of the aerogels are determined as follows: %Vs =



1−

Va Vg



×

100

Pore volume (cm3 /g) =



Porosity = 1 −

b s



×



(4) 1 b

100

−

1 s



(5)

(6)

Va and Vg are the volume of aerogel and alcogel respectively. s

is the skeletal and, b is the bulk density of the silica aerogel. 3.3.

Optical properties

In many procedures, the resulted silica aerogels are transparent. This is an unusual property for a porous material. The reason for this relatively rare combination of traits arises because the aerogel microstructure has a scale small compared to the wavelength of light. Transparency occurs because there is a small amount of scattering in the visible, the scattered light has a relatively isotropic angular distribution, and exhibits little multiple scattering (Russo and Hunt, 1986; Kamiuto et al., 1993; Hunt and Berdahl, 1984; Beck et al., 1989). This behavior can be described by Rayleigh scattering theory. Rayleigh scattering is characterized by the isotropic scattering of vertically polarized incident light, an intensity that varies with scattering angle as cos 2 Â for horizontally polarized inci-

  journal of materials processing technology 199 (2008) 10–26

dent lightand a wavelength dependence of scattered lightthat varies as1/4 . Aerogels are unlike Rayleigh scatterers because they also exhibit a wavelength independent component of  scattering that may not be isotropic and some samples can deviate significantly from Rayleigh angular distribution (Hunt and Berdahl, 1984). In many applications, their transparency plays an important role. This remark attracted the attention of scientists working in high energy physics with charged particles and looking for the construction of a new type of a solid-state Cerenkov counter (Yokogawa and Berdahl, 1995). Due to their very high thermal insulation properties and their optical transparency in the visible region, they were proposed for double plane windows (  Jensen et al., 2004; Schultz et al., 2005b). Thus, multiple efforts have been undertaken so as to improve this characteristic of silica aerogels. Initially, various physically oriented approaches where taken for solving this problem, such as investigation of the influence of the drying process, of water adsorbed to Si–OH groups, or of  absorbed organic components. Heating the aerogels improves their transparency due to desorption of water and burning of  organic components (Buzykaev et al., 1999). Also, the sol–gel processparameters and the typeof silation agent greatly influence the optical properties of aerogels ( Rao et al., 2001). It has been shown that the two-step synthetic method resulted in more transparent aerogels than those obtained by onestep synthesis (Danilyuk et al., 1999). Bulk scattering can be minimized by the selection of optimal synthesis parameters. Recently, Adachi et al. (2005) synthesized new aerogels with a refractive index larger than 1.03 by introducing di-methylformamide (DMF), as solvent in sol–gel process. A review of the works devoted to this problem was published by Pajonk (1998). Refractive index (n) ofthe aerogelswas calculated with the formula of Clausius-Mosotti (Eq. (8)), (Poelz and Riethmuller, 1982). n=

3  n2s − 1 2 s n2s + 2

=

1 + 0.19

(7)

where n and ns are refractive indices and  and s the densities of aerogel and silica, respectively. Using ns =1.46 and s =2.2g/cm 3 in the equation, n =1+0.19 will be obtained. 3.4.

23

sample, given as the product of the geometrical thickness and the optical extinction coefficient of the aerogel. At low temperatures, the radiative component of thermal transport is low, but at higher temperatures, radiative transport becomes a dominant mode of thermal conduction. An attempt to calculatethe total thermal conductivity arisingfrom thesum of these three modes canprove to be difficult because the modes are coupled .For measurement of the thermal conductivity, a Vacuum Insulation Conductivity Tester – VICTOR – may be employed. Some authors reported the use of  a hot wire method (Kun-Hong Lee et al., 1995). 3.5.

Hydrophobicity

Silica aerogels can either be hydrophilic or hydrophobic, depending on the conditions during synthesis. The silanol polar groups Si–OH present in the aerogel structure are the main source of hydrophilicity because they can promote the adsorption of water. Generally, aerogels synthesized by unmodified hydrolysis and condensation of alkylorthosilicates and dried by high temperature SCD are hydrophobic, and those dried by CO 2 are hydrophilic. This difference is due to the different surface groups formed during the SCD process. LTSCD results in hydroxyl groups (–OH) on the surface resulting hydrophilic aerogels. HTSCD allows for the reaction of the surface hydroxyl groups with the solvent to form methoxy groups (–OCH 3 )X and thus results in hydrophobic aerogels. Fourier transform infrared spectroscopy (FTIR) and NMR was employed to investigate the chemical bonding state of  aerogels (El Rassy and Pierre, 2005). There are two different routes to increase the hydrophobicity of an aerogel: The hydrophobic character can be increased by the addition of a silylating agent during the sol–gel step. This principle is used in APD methods ( Deshpande, 1996). A secondapproach for increasing thehydrophobicity is the modification of theaerogel surface after drying. Thesurfaceof  hydrophilic aerogels can be modified by reaction with gaseous methanol (Kun-Hong Lee et al., 1995). The hydrophobicity of the aerogels was tested by measuring the contact angle ( Â ), of a water droplet with the aerogel surface using the formula,

Thermal conductivity

Because of porosity and nanometer pore size, silica aerogels are highly insulating materials with thermal conductivity lower thanstill air. Kistler demonstratedthat the thermal conductivity of an aerogel is in the order of 0.02 W/mK at ambient pressure in air and in the order of 0.01 W/mK when evacuated (Kistler, 1932). Because silica aerogels have a very small ( ∼1–10%) fraction of solid silica, thus exhibit a lower solid conductivity and hence, transmit a lower thermal energy. Gases are also able to transport thermal energy through the aerogel. The pores of silica aerogel are open and allow the passage of gas through the material. The final mode of thermal transport through silica aerogels involves infrared radiation. An important parameter that influences this transfer route is the optical thickness of the

Â

=

2tan−1

2h

  w

(8)

where h is the height and w is the width of the water droplet touching the aerogel surface. Traveling moicroscopes were used for measuring h and w (Fig. 13).The surface modification was confirmed with the Fourier transform infrared spectroscopy (FTIR) using an IR spectrophotometer. The thermal stability of aerogels in terms of retention of hydrophobicity was estimated from the thermogravimetric and differential thermal analyses (TGA–DTA) as well as by heating the aerogels in the furnace and then putting the heated samples on the water surface. The retention of  hydrophobicity (water-repelling property) was judged from the absorption of water by the aerogels. Hydrophobic gels exhibit hydrophobicity only for a certain period of time. Being

24

j o u r n a l o f m a t e r i a l s p r o c e s s i n g t e c h n o l o g y 1 9 9 ( 2 0 0 8 ) 10–26

Fig. 13 – Photograph of water droplet on the modified silica aerogels.

exposed to air over a long time resulted in adsorbing water, a characteristic that is not typical of hydrophobic materials (Schwertfeger et al., 1996).

4.

Conclusion

This paper provides a comprehensive review of the synthesis, structure, properties and characterization of silica aerogels. Aerogels show great promise for use in variety of technological areas where special structure and physical properties are required. Substantial progress have has been made in the development, processing and characterization of aerogel materials over the recent years. Special attention has been paid to theuse of inexpensive precursors such as sodiumsilica (waterglass) andthe drying technology to makethe production commercial. Silica aerogel synthesis with various materials and process conditions and also the properties and method of determination are reviewed and summarized in this paper.

references

Adachi, I., Sumiyoshi, T., Hayashi, K., Iida, N., Enomoto, R., Tsukada, K., Suda, R., Matsumoto, S., Natori, K., Yokoyama, M., Yokogawa, H., 1995. Nucl. Instr. Meth. Phys. Res. A 355, 390. Adachi, I., Fratina, S., Fukushim, T., Gorisek, A., Iijima, T., Kawai, H., Konishi, M., Korpar, S., Kozakai, Y., Krizan, P., Matsumoto, T., Mazuka, Y., Nishida, S., Ogawa, S., Ohtake, S., Pestotnik, R., Saitoh, S., Seki, T., Sumiyoshi, T., Tabata, M., Uchida, Y., Unno, Y., Yamamoto, S., 2005. Nucl. Instr. Meth. Phys. Res. A 553, 146–151. Alie, C., Pirard, R., Lecloux, A.J., Pirard, J.P., 1999. J. Non-Cryst. Solids 246, 216. Alie’, C., Ferauche, F., Pirard, R., Lecloux, A.-J., Pirard, J.-P., 2001a. J. Non-Cryst. Solids 289, 88. Alie’, C., Pirard, R., Lecloux, A.-J., Pirard, J.-P., 2001b. J. Non-Cryst. Solids 285, 135. Alie’, C., Pirard, R., Pirard, J.-P., 2002. J. Non-Cryst. Solids 301, 304. Anappara, A.A., Rajeshkumar, S., Mukundan, P., Warrier, P.R.S., Ghosh, S., Warrier, K.G.K., 2004. Acta Mater. 52, 369. Anderson, M.T., Sawyer, P.S., Rieker, T., 1998. Micropor. Mesopor. Mater. 20, 53.

Asner, D., Butler, F., Dominick, J., Fadeyev, V., Masek, G., Nemati, B., Skubic, P., Strynowski, R., 1996. Nucl. Instr. Meth. Phys. Res. A374, 286. Attia, Y.A., Ahmed, M.S., Zhu, M., 1994. In: Attia, Y.A. (Ed.), Sol–Gel Processing and Applications. Plenum, New York, p. 311. Barbieri, O., Ehrburger-Dolle, F., Rieker, T.P., Pajonk, G.M., Pinto, N., Rao, A.V., 2001. J. Non-Cryst. Solids 285, 109. Beck, Caps, R., Fricke, J., 1989. J. Phys. D 22, 730. Bhagat, S.D., Kim, Y.-H., Ahn, Y.-S., Yeo, J.-G., 2006. Microporous Mesoporous Mater. 96, 237–244. Bhagat, S.D., Kim, Y.-H., Ahn, Y.-S., Yeo, J.-G., 2007. Applied Surface Science 253, 3231–3236. Bisson, A., Rodier, E., Rigacci, A., Lecomte, D., Achard, P., 2004. J. Non-Cryst. Solids 350, 230–237. Brinker, C.J., Sherer, G.W., 1990. The Physics and Chemistry of  Sol-Gel Processing. Acad. Press, New York. Brinker, C.J., Keefer, K.D., Schaefer, D.W., Assink, R.A., Kay, B.D., Ashley, C.S., 1984. J. Non-Cryst. Solids 63, 45. Bryans, T.R., Brawner, V.L., Quitevis, E.L., 2000. J. Sol-Gel Sci. Technol. 17, 211. Buzykaev, A.R., Danilyuk, A.F., Ganzhur, S.F., Kravchenko, E.A., Onuchin, A.P., 1999. Nucl. Instr. Meth. Phys. Res. A 433, 396. Cao, B., Zhu, C., 1999. J. Non-Cryst. Solids 246, 34. Carraher Jr., C.E., 2005. Polym. News 30 (12), 386. Chou, K., Lee, B.I., 1994. J. Mater. Sci. 29, 3565. Colon, L.A., Guo, Y., Fermier, A., 1996. Anal. Chem. 68, 2753. Constantin, S., Freitag, R., 2000. J. Chromatogr. A 887, 253. Cooper, D.W., 1989. Part. Sci. Technol. 7, 371. Dai, S., Ju, Y.H., Gao, H.J., Lin, J.S., Pennycook, S.J., Barnes, C.E., 2000. Chem. Commun., 243. Danilyuk, A.F., Kravchenko, E.A., Okunev, A.G., Onuchin, A.P., Shaurman, S.A., 1999. Nucl. Instr. Meth. Phys. Res. A 433, 406–407. Davis, P.J., Brinker, C.J., Smith, D.M., 1992a. J. Non-Cryst. Solids 142, 189. Davis, P.J., Brinker, C.J., Smith, D.M., Assink, R.A., 1992b. J. Non-Cryst. Solids 142, 197. DeLeo, R., Lagamba, L., Manzari, V., Nappi, E., Scognetti, T., Alemi, M., Becker, H., Forty, R., Adachi, I., Suda, R., Sumiyoshi, T., Leone, A., Perrino, R., Matteuzzi, C., Seguinot, J., Ypsilantis, T., Cisbani, E., Frullani, S., Garibaldi, F., Iodice, M., Uriuoli, G.M., 1997. Nucl. Instr. Meth. Phys. Res A 401, 187. Deng, Z., Wang, J., Wei, J., Shen, J., Zhou, B., Chen, L., 2000. J. Sol-Gel Sci. Technol. 19, 677–680. Deshpande, R., Smith, D., Brinker, C.J. US Patent 5,565,142 issued 15.10.1996. Deshpande, R., Hua, D.W., Smith, D.M., Brinker, C.J., 1992. J. Non-Cryst. Solids 144, 32. Dieudonne, P., Hafidi Alaoui, A., Delord, P., Phalippou, J., 2000. J. Non-Cryst. Solids 262, 155. Dunn, B.C., Cole, P., Covington, D., Webster, M.C., Pugmire, R.J., Ernst, R.D., Eyring, E.M., Shah, N., Huffman, G.P., 2004. J. Non-Cryst. Solids 350, 23–30. Ehrburger-Dolle, F., Dallamano, J., Elaloui, E., Pajonk, G., 1995. J. Non-Cryst. Solids 186, 9. Einarsrud, M.-A., 1998. J. Non-Cryst. Solids 225, 1. Einarsrud, M.A., Nilsen, E., 1998. J. Non-Cryst. Solids 226, 122. Einarsrud, M.-A., Kirkedelen, M.B., Nilsen, E., Mortensen, K., Samseth, J., 1998. J. Non-Cryst. Solids 231, 10. Einarsrud, M.-A., Nilsen, E., Rigacci, A., Pajonk, G.M., Buathier, S., Valette, D., Durant, M., Chevalier, B., Nitz, P., Ehrburger-Dolle, F., 2001. J. Non-Cryst. Solids 285, 1. El Rassy, H., Pierre, A.C., 2005. J. Non-Cryst. Solids 351, 1603–1610. Emmerling, A., Wang, P., Popp, G., Beck, A., Fricke, J., 1993. Proceedings of IX International Conference on Small Angle Scattering, Saclay Universita W ¨urzburg. Estella, J., Echeverria, J.C., Laguna, M., Garrido, J.J., 2007. Microporus Mesoporous Mater. 102, 274–282. Fricke, J., 1988. J. Non-cryst. Solids 100, 169.

  journal of materials processing technology 199 (2008) 10–26

Fricke, J., Emmerling, A., 1992. J. Am. Ceram. Soc. 75, 2027. Fricke, J., Tillotson, T., 1997. Thin Solid Films 297, 212. Gauthier, B.M., Bakrania, S.D., Anderson, A.M., Carroll, M.K., 2004.  J. Non-Cryst. Solids 350, 238–243. Gesser, H.D., Goswami, P.C., 1989. Chem. Rev. 89, 765–788. Gimel, J.C., Durand, D., Nicolai, T., 1995. Phys. Rev. B 51, 1134. Glatter, O., Kratky, O., 1982. Small Angle X-ray Scattering. Academic Press, London. Gross, J., Coronado, P., Hrubesh, L., 1998. J. Non-Cryst Solids 225, 282. Guo, L., Hycon Lee, J., Beaucage, G., 1999. J. Non-Cryst. Solids 243, 61. Hæreid, S., Anderson, J.M., Einarsrud, M.-A., Hua, D.W., Smith, D.M.J., 1995. Non-Cryst. Solid 185, 221. Haereid, S., Dahle, M., Lima, S., Einarsrud, M.A., 1995a. J. Non-Cryst. Solids 186, 96. Haereid, S., Nilsen, E., Einarsrud, M.-A., 1995b. J. Porous Mater. 2, 315. Haereid, S., Nilsen, E., Einarsrud, M.A., 1996. J. Non-Cryst. Solids 204, 228. Hafidi Alaoui, A., Woignier, T., Phalippou, J., Scherer, G.W., 1998. J. Sol-Gel Sci. Technol. 13, 365. Harreld, J.H., Ebina, T., Tsubo, N., Stucky, G., 2002. J. Non-Cryst. Solids 298, 241. Hdach, H., Woignier, T., Phalippou, J., Scherer, G.W., 1990. J. Non-Cryst. Solids 121, 202. Hrubesh, L.W., 1998. J. Non-Cryst. Solids 225, 335. Hua, D.W., Anderson, J., Di Gregorio, J., Smith, D.M., Beaucage, G., 1995. J. Non-Cryst. Solids 186, 142. Hunt, A.J., 1998. J. Non-Crystal. Solids 225, 303–306. Hunt, A.J., Berdahl, P., 1984. Mater. Res. Soc. Symp. Proc. 32, 275. H ¨using, N., Dissertation, W ¨urzburg 1997. Husing, N., Schubert, U., 1998. Angew. Chem. Int. Ed. 37, 22. Hwang, S.-W., Jung, H.H., Hyun, S.H., Ahn, Y.S., 2007. J. Sol-Gel Sci. Technol. 41, 139–146. Ishino, M., Chiba, J., En’yo, H., Funahashi, H., Ichikawa, A., Ieiri, M., Kanda, H., Masaike, A., Mihara, S., Miyashita, T., Murakami, T., Nakamura, A., Naruki, M., Muto, R., Ozawa, K., Sato, H.D., Sekimoto, M., Tabaru, T., Tanaka, K.H., Yoshimura, Y., Yokkaichi, S., Yokoyama, M., Yokgawa, H., 2001. Nucl. Instr. Meth. Phys. Res. A 457, 581. Ishizuka, N., Minakuchi, H., Nakanishi, K., Soga, N., Hosoya, K., Tanaka, N., 1998. J. Chromatogr. A 797, 133. Ishizuka, N., Minakuchi, H., Nakanishi, K., Soga, N., Hosoya, K., Tanaka, N., 2000. Anal. Chem. 72, 1275.  Jensen, K.I., Schultz, J.M., Kristiansen, F.H., 2004. J. Non-Cryst. Solids 350, 351–357.  Judenstein, P., Titman, J., Stamm, M., Schmidt, H., 1994. Chem. Mater. 6, 127. Kamiuto, K., Saitoh, S., Tokita, Y., 1993. J. Quant. Spectrosc. Radiant. Transfer 50, 293. Karmakar, B., De, G., Ganguli, D., 2000. J. Non-Cryst. Solids 272, 11. Kim, G.S., Hyun, S.H., 2001. Ceramist 4, 5. Kim, G.S., Hyun, S.H., 2003. J. Mater. Sci. 38, 1961. Kim, G.S., Hyun, S.H., 2004. Thin Solid Films 460, 190–200. Kirkbir, F., Murata, H., Meyers, D., Ray Chaudhuri, S., Sarkkar, A., 1996. J. Sol-Gel Sci. Technol. 6, 203. Kirkbir, F., Murata, H., Meyers, D., Ray Chaudhuri, S., 1998. J. Non-Cryst. Solids 225, 14. Kistler, S.S., 1932. J. Phys. Chem. 36, 52. Kocon, L., Despetis, F., Phalippou, J., 1998. J. Non-Cryst. Solids 225, 96–100. Komarneni, S., Roy, R., Selvaraj, U., Malla, P.B., Breval, E., 1993. J. Mater. Res. 8, 3163. Lee, K.-H., Kim, S.-Y., Yoo, K.-P., 1995. J. Non-Cryst. Solids 186, 307. Land, V., Harris, T., Teeters, D., 2001. J. Non-Cryst. Solids 283, 11. Lee, C.J., Kim, G.S., Hyun, S.H., 2002. J. Mater. Sci. 37, 2237–2241. Liu, C., Komarneni, S., 1995. In: Komarneni, S., Smith, D.M., Beck,  J.S. (Eds.), Advances in Porous Materials, Mater. Res. Soc.

25

Symp. Proc. 371. Materials Research Society, Pittsburgh, p. 217. Loy, D.A., Russick, E.M., Yamanaka, S.A., Baugher, B.M., 1997. Chem. Nat. 9, 2264. MacKenzie, J.D., 1988. J. Non-Cryst. Solids 100, 162. Mansur, H.S., Vasconcellos, W.L., Lenza, R.S., Orefice, R.L., Reis, E.F., Lobato, Z.P., 2000. J. Non-Cryst. Solids 273, 109. Martin, J., Hosticka, B., Lattimer, C., Norris, P.M., 2001. J. Non-Cryst. Solids 288, 222. Mezza, P., Phalippou, J., Sempere, R., 1999. J. Non-Cryst. Solids 243, 75. Minakuchi, H., Nakanishi, K., Soga, N., Ishizuka, N., Tanaka, N., 1996. Anal. Chem. 68, 3498. Minakuchi, H., Nakanishi, K., Soga, N., Ishizuka, N., Tanaka, N., 1997. J. Chromatogr. A 762, 121. Mincov, I., Petkov, M.P., Tsou, P., Troev, T., 2004. J. Non-Cryst. Solids 350, 253–258. Moner-Girona, V, Roig, A., Molins, E., Llibre, J. Materials of ISA6, Albuquerque, N.M., USA October 8–11, 2000. Moner-Girona, M., Roig, A., Molins, E., 2003. J. Sol-Gel Sci. Technol. 26, 645–649. Murata, H., Meyers, D.E., Kirkbir, F., Ray Chaudhuri, S., Sarkar, A., in: J.D. Mackenzie (Ed.), Sol-Gel Optics, vol. III, SPIE – The International Society for Optical Engineering Washington, 1994, vol. 2288, p. 709. Nakanishi, K., Minakuchi, H., Soga, N., Tanaka, N., 1997. J. Sol-Gel Sci. Technol. 8, 547. Nakanishi, K., Minakuchi, H., Soga, N., Tanaka, N., 1998. J. Sol-Gel. Sci. Technol. 13, 163. Nicolaon, G.A., Teichner, S.J., 1968. Bull. Soc. Chim., 1906. Novak, Z., Knez, Z., Hadolin, M., 1999. Second European Congress of Chemical Engineering. Montpellier. Nyquist, R.A., Kagel, R.O., 1997. Handbook of Infrared and Raman Spectra of Inorganic Compounds and Organic Salts, vol. 4. Academic Press, Inc, San Diego, p. 94. Pajonk, G.M., 1998. J. Non-Cryst. Solids 225, 307. Pajonk, G.M., Teichner, S.J., 1985. In: Fricke, J. (Ed.), Proceedings of  the First International Symposium on Aerogels. Springer, Berlin, p. 193. Parvathy Rao, A., Pajonk, G.M., Rao, A.V., 2005. J. Mater. Sci. 40, 3481–3489. Pierre, A.C., Pajonk, G.M., 2002. Chem. Rev. 102, 4243. Platzer, W.J., Bergkvist, M., 1993. Solar Energy Mater. Solar Cells 31, 243. Poco, J.F., Coronado, P.R., Pekala, R.W., Hrubesh, L.W., 1996. Mater. Res. Soc. Symp. 431, 297. Poelz, G., Riethmuller, R., 1982. Nucl. Instrum. Methods 195, 491. Pope, E.J.A., Mackenzie, J.D., 1986. J. Non-Cryst. Solids 87, 185. http://www-pat.llnl.gov/H Div/Positrons/PositronMaterials.html. Posselt, D., 1991. Structural Studies of Silica-Aerogel – A Mass–Fraktal Model System. Ris-National Laboratory, Roskilde. Puska, M.J., Nieminen, R.M., 1994. Rev. Mod. Phys. 66, 841. Rao, A.V., Bhagat, S.D., 2004. Solid State Sci. 6, 945. Rao, A.V., Haranath, D., 1999. Micropor. Mesopor. Mater. 30, 267. Rao, A.V., Kulkarni, M.M., 2002. Mater. Res. Bull. 306, 1667. Rao, A.V., Kulkarni, M.M., 2003. Mater. Chem. Phys. 77, 819. Rao, A.V., Wagh, P.B., 1998. Mater. Chem. Phys. 53, 13–18. Rao, A.V., Wagh, P.B., Haranath, D., Risbud, P.P., Kumbhare, S.D., 1999. Ceram. Int. 25, 505. Rao, A.V., Nilsen, E., Einarsrud, M.A., 2001. J. Non-Cryst. Solids 296, 165–171. Rao, A.V., Kulkarni, M.M., Amalnerkar, D.P., Seth, T., 2003. J. Non-cryst. Solids 330, 187. Rao, A.V., Parvathy Rao, A., Kulkarni, M.M., 2004a. J. Non-Cryst. Solids 350, 224–229. Rao, A.V., Pajonk, G.M., Bhagat, S.D., Barboux, P., 2004b. J. Non-Cryst. Solids 350, 216–223.

26

j o u r n a l o f m a t e r i a l s p r o c e s s i n g t e c h n o l o g y 1 9 9 ( 2 0 0 8 ) 10–26

Rao, A.P., Pajonk, G.M., Rao, A.V., 2005a. J. Mater. Sci. 40, 3481–3489. Rao, A.P., Rao, A.V., Pajonk, G.M., 2005b. J. Sol Gel Sci. Technol. 36, 285. Rao, A.V., Bhagat, S.D., Hirashima, H., Pajonk, G.M., 2006. J. Colloid Interf. Sci. 300, 279–285. Rao, A.V., Hegde, N.D., Hirashima, H., 2007a. J. Colloid. Interf. Sci. 305, 124–132. Rao, A.P., et al., 2007b. Appl. Surf. Sci. 253, 6032–6040. Reetz, M.T., Zonta, A., Simpelkamp, J., 1996. Biotechnol. Bioeng. 49, 527. Reichenauer, G., 2004. J. Non-Crystal. Solids 350, 189–195. Reichenauer, G., Scherer, G.W., 2001. J. Non-Cryst. Solids 285, 167. Richard Brundle, C., Evans Jr., C.A., 1992. Encyclopedia of  Materials Characterization. Reed Publishing (USA) Inc. Rodriguez, S.A., Colon, L.A., 1999. Anal. Chim. Acta 397, 207. Russo, R.E., Hunt, A.J., 1986. J. Non-Cryst. Solids 86, 219. Schadl, J., Mersmann, A.B., 1985. In: Mujumdar, A.S. (Ed.), Proc. Intl. Drying Symposium, pp. 170–177. Scherer, G.W., Smith, D.M., 1995. J. Non-Cryst. Solids 189, 197. Scherer, G.W., Smith, D.M., Qiu, X., Anderson, J.M., 1995. J. Non-Cryst. Solids 186, 316. Scherer, G., Hæreid, S., Nilsen, E., Einarsrud, M.-A., 1996. J. Non-Cryst. Solids 202, 42. Scherer, G.W., Gross, J., Hrubesh, L.W., Coronado, P.R., 2002. J. Non-Cryst. Solids 311, 259. Schmidt, M., Schwertfeger, F., 1998. J. Non-Cryst. Solids 225, 364. Schneider, M., Baiker, A., 1995. Catal. Rev. -Sci. Eng. 37, 515. Schultz, P.J., Lynn, K.G., 1988. Rev. Mod. Phys. 60, 701. Schultz, J.M., Jensen, K.I., Kristiansen, F.H., 2005a. Sol. Energy Mat. Sol. Cells. 89, 275. Schultz, J.M., Jensen, K.I., Kristiansen, F.H., 2005b. Solar Mater. Solar Cells 89, 275–285. Schuth, F., Sing, K.S.W., Weitkamp, J. (Eds.), Hand Book of Porous Solids, vol. 3 (2002) p. 2014. Schwertfeger, F., Zimmermann, A., Krempel, H., WO 96/25850, issued 1996. Schwertfeger, F., Frank, D., Schmidt, M., 1998. J. Non-Cryst. Solids. 225, 24–29. Sharp, K. WO 93/23333, issued 1993. Shi, F., Wang, L., Liu, J., 2006. Mater. Lett. 60, 3718–3722. Sing, K.S.W., Everett, D.H., Haul, R.A.W., Moscou, L., Pierotti, R.A., Rouquerol, J., Siemieniewska, T., 1985. Pure Appl. Chem. 57, 603.

Smith, D.M., Stein, D., Anderson, J.M., Ackermn, W., 1995. J. Non-Cryst. Solids 186, 104. Smitha, S., Shajesh, P., Aravind, P.R., Rajesh Kumar, S., Krishna Pillai, P., Warrier, K.G.K., 2006. Microporus Mesoporous Mater. 91, 286–292. Stauffer, D., Aharony, A., 1994. Introduction to percolation theory, revised seconnd ed. Taylor and Francis, London. Strøm, R.A., Masmoudi, Y., Rigacci, A., Petermann, G., Gullberg, L., Chevalier, B., Einarsrud, M.-A., 2007. J. Sol-Gel Sci. Technol. 41, 291–298. Suh, D.J., Park, T.J., Sonn, J.H., Lim, J.C., 1999. J. Mater. Sci. Lett. 18, 1473. Sumiyoshi, T., Adachi, I., Enomotoi, R., Iijima, T., Suda, R., Yokoyama, M., Yokogawa, H., 1998. J. Non-Cryst. Solids 225, 369. Tajiri, K., Igarashi, K., 1998. Solar Energy Mater. Solar Cells 54, 189. Tang, Q., Wang, T., 2005. J. Supercrit. Fluids 35, 91–94. Tewari, P.H., Hunt, A.J., Lofftus, K.D., 1985. Mater. Lett. 3, 363. Tewari, P.H., Hunt, A.J., Lofftus, K.D., 1986. Springer Proc. Phys. 6, 31. Titulaer, M.K., Jansen, J.B.H., Geus, J.W., 1994. J. Non-Cryst. Solids 170, 11. Tyler, L.J., Dow Corning, GB 682574, 1959. Tyler, L.J., Dow Corning, US 3015645, 1962. Ulrich, D.R., 1990. J. Non-Cryst. Solids 121, 465. Unsulu, B., Sunol, S., Sunol, A., 2001. J. Non-Cryst. Solids 279, 110. van Bommel, M.J., de Haan, A.B., 1995. J. Non-Cryst. Solids 186, 78. Vollet, D.R., Donatti, D.A., Ibanez Ruiz, A., 2001. J. Non-Cryst. Solids 288, 81. Wagh, P.B., Ingale, S.V., 2002. Ceram. Int. 28, 43–50. Wagh, P.B., Begag, R., Pajonk, G.M., Venkasteswara Rao, A., Haranath, D., 1999. Mater. Chem. Phys. 57, 214. Wawrzyniak, P., Rogacki, G., Pruba, J., Bartczak, Z., 2001. J. Non-Cryst. Solids 285, 50–56. Woignier, T., Phalippou, J., 1987. J. Non-Cryst. Solids 93, 17. Wu, G., Wang, J., Shen, J., Yang, T., Zhang, Q., Zhou, B., Deng, Z., Bin, F., Zhou, D., Zhang, F., 2000. J. Non-Cryst. Solids 275, 169. Yoda, S., Ohshima, S., 1999. J. Non-Cryst. Solids 248, 224. Yokogawa, M., Yokoyama, 1995. J. Non-Cryst. Solids 186, 23. Zhang, X., Huang, S., 2001. J. Chromatogr. A 910, 13. Zhou, B., Shen, J., Yuehua, W., Wu, G., Ni, X., 2007. Mater. Sci. Eng. C 27, 1291–1294.