Silica Glass from Aerogels

by Michel Prassas

Aerogel evolution during sintering

The problem of monolithic gels Two technologically important processing steps were conditioning the successful formation of a glass from a gel: Drying for evacuate the  chemicals (mutual solvent of water and alkoxy silane, and reaction byproducts) present within the gel network and, sintering to densify the gel by eliminating the porosity left after drying without destroying the amorphous structure.

Browse Article Glass from gels Chemistry Hypercritical drying Gel to glass transformation Conclusions References & Links

In order however to preserve the expected sol-gel advantages, it was obvious that gels should be obtained in a monolithic form of appreciable size and maintain their integrity throughout all subsequent processing steps (drying, sintering ) which transform a gel to a glassy material. 

Freshly prepared silica gels contain an appreciable amount of solvent (usually 70 to 90 wt %) which must be eliminated. 
The solid silica network formed by hydrolysis and polycondensation of Si alkoxides is made up by Silica species of a few tens of nanometer size.
Capillary stress appears when the liquid move inside the pores during drying and form a liquid-gas curve interface. 

The liquid inside the pores (which size is similar to ultimate silica particles) exert during drying a stress in the "walls" of the capillaries which is inversely proportional to the pores diameter (see figure 3).  
In the case of an alcoholic solvent (i.e. methanol, surface tension g = 0.0022 Nm^-1) and for a 10 nm pore radius, the capillary stress given by Laplace's law (figure 3) is of  the order of more than 10 ⁶ N/m². 

This is high enough to break the gel into useless pieces of a few mm³ in size. 
The first attempts by Yamane in Japan and Yoldas in USA to keep the integrity of the gels, took weeks and even months of a careful drying procedure under experimental conditions which most of the time were extremely difficult to reproduce (a number of pin holes on the top of the gel vessel...) but had the merit to demonstrate that obtaining monolithic gels from alkoxides was difficult but realistic. 

The size of these gels never exceeded  2 to 4 cm diameter disks with 1 to 2 cm in thickness. 

A variety of techniques have been suggested (solvent exchange or use of surfactants to minimize surface tension effects, gel aging to reinforce the silica network, in situ chemical modification of the capillary surface etc.) None of the investigated techniques were able to reproducibly provide large size, cracks free dry monoliths. 

In 1931 however Samuel Kistler a genius researcher at the University of Illinois had demonstrated practically what thermodynamics teach us. Above its critical pressure and temperature any substance is present only in one phase neither liquid nor gas. No more two phase and therefore no reason for capillary forces to appear. 
This principle was for many years since Kistler used and greatly improved by researchers mainly in the catalysis field to produce high surface area catalysts of practically any oxide. 

The first attempt of using hypercritical drying to obtain monolithic piece of porous silica, as glass precursors,  was made at the Glass laboratory of the University of Montpellier, France by two researchers M. Prassas and J. Phalippou in 78.  The technique used previously by Nicolaon et Teichner at the University of Lyon to elaborate high surface area catalysts from alkoxides based gels was improved and adapted to easy produce monoliths of more than 30 cm in size, a dimension which today seems ridiculous but at that time was a significant improvement over the existing methods.

Hypercritical drying. 

The typical apparatus used to obtain a monolithic gel by hypercritical solvent evacuation is illustrated below:

An autoclave from inoxidized steel is used to bring the initial solution (in this case  Si(OCH3)4 - CH3OH - H2O) above the critical point of the solvent (CH3OH/H2O) present within the pores of the silica gel. To avoid crossing the liquid-gas equilibrium an additional amount of solvent is added inside the autoclave and the system  allowed to reach and exceed the critical point (for pure methanol  T_cr = 240°C , P_cr = 79.7 bars). As soon as the critical temperature is reached the vapors of the solvent are slowly evacuated, keeping the temperature constant. When the pressure reach atmospheric pressure, the autoclave is flushed with dry Argon and then cooled to room temperature. The entire operation (sol preparation, heating, solvent evacuation, cooling take less than 10 h.) 

The  produced porous gel so-called aerogel (from the greek aera = air )  is monolithic, extremely fragile, it shows no shrinkage  and is one of the lightest solid materials in earth. Typical density of silica aerogels is  in the range of 0.1 to 0.3 g/cc.  
A variety of glass compositions can be synthesized and dried under  hypercritical conditions (see figure 5).   

Later, the procedure was further improved  with the use of CO₂ (which certainly needs solvent exchange but is much more safe than the alcoholic hypercritical evacuation). For more details see the Aerogel web site at Laurence Berkeley National Laboratory