In a previous post, we have already written about the mechanical stability of cellular glass. We even have discussed the static fatigue limit of glass and so cellular glass. These posts, based on acoustic emission experiments and self organized criticality (1988) take into account the interaction between the different cells. In case a cell breaks, the load is redistributed over the neighbor cells.

In a 1981 paper of the NASA, the mechanical stability of cellular glass is experimentally and theoretically studied. I give the abstract in the following:

Cellular glasses are prime candidate materials for the structural substrate of mirrored glass for solar concentrator reflecting panels. These materials possess properties desirable for this application such as high stiffness to weight ratio, dimensional stability, projected low cost in mass production and, importantly, a close match in thermal expansion coefficient with that of the mirror glass. These materials are brittle, however, and susceptible to mechanical failure from slow crack growth caused by a stress corrosion mechanism.
This report details the results of one part of a program established to develop improved cellular glasses and to characterize the behavior of these and commercially available materials. Commercial and developmental cellular glasses were tested and analyzed using standard testing techniques and models developed from linear fracture mechanics. Two models describing the fracture behavior of these materials are developed. Slow crack growth behavior in cellular glass was found to be more complex than that encountered in dense glasses or ceramics. The crack velocity was found to be strongly dependent upon water vapor transport to the tin of the moving crack. The existence of a
static fatigue limit was not conclusively established, however, it is speculated that slow crack growth behavior in Region I may be slower, by orders of magnitude, than that found in dense glasses.

The motivation of NASA was already described in a previous post and it even already applied to replace concrete. On top of that, cellular glass and glass have the same thermal expansion coefficient if the foam is made from the same glass. This makes cellular glass as the preferred support for mirrors.

The mechanical stability was measured by measuring the velocity of a single crack under a load. In a previous post, we described how the acoustic emisson detects the interaction of many micro cracks. In fact, these microcracks are a complex system inducing small and large events. We prefer this method and not the study of a single macroscopic crack. The acoustic emission technique allows to measure the static fatigue limit of cellular glass.