The fracture strengths of the foam glass are significantly determined by the properties of its starting materials (generally glass and an expanding agent) as well as the chemical-physical processes during the production process. In this regard the following important relationships should be taken into account:

• 1)the chemical condition of the glass raw material;
• 2)the type of expanding agent;
• 3)the mass composition of the starting mixture of glass powder, expanding agent and possibly an additive;
• 4)the quantity of the chemically unreacted expanding agent in the cell wall films;
• 5)the specific surface, homogeneity and grain distribution curve of the starting mixture; and
• 6)the management of the temperature profile line during the production process.

Residual internal stresses, which arise during the cooling process of the foam glass (annealing process) promote brittle fracture and reduce the final strengths. For this reason, the influence of the annealing process on the fracture strengths of the foam glass should be given a high degree of attention. During the production of the foam glass tested in this work, the cooling process was controlled in such a manner that the internal stresses as a result of annealing could be neglected.

In section No. 1.3.1 of chapter A it was demonstrated using the example of a spherical foam glass sample made of sintered coloured waste glass powder that traces of cristobalite may be present in the base material of the foam glass. This is α - cristobalite, which is formed during the cooling of the foam glass

in the transition range of 220°C to 280°C from β - cristobalite. Since this crystal structure conversion is connected with a large concurrent volume reduction, shrinkage stresses are caused which have a negative influence on the fracture strengths of the foam glass. For this reason the cooling process of the glass foam samples produced by us was controlled in such a way that the formation of cristobalite with its negative influence on the fracture strengths of the end product could be ignored.

Regarding the influence of the microstructure of the foam glass on its mechanical behaviour, has already been extensively reported in this paper. Therefore it now only remains to be added that an influence of the static properties of the polyhedron foam glass due to the content of open cells can be safely disregarded.

The dependency of the fracture- and elastic behaviour of a polyhedron foam glass due to its structure can be expressed using the relationship  using the example of the test cylinder (see Fig. No. 49). At large  L̅_{3 / L} at points where the cell wall films of the polyhedron foam glass contact the test cylinder surface, strong edge effects occur, whereas in the opposite case (small ( L̅_{3 / L} klein) they can be neglected.

As already mentioned, with the compressive tests the test pieces were supported in the test machine using rubber plates. At the end areas of the test piece this support causes a favourable spatial compressive stress condition. It could also be that ceteris parabus a reducing mean sectional length L̅₃ increases the fracture strengths of the polyhedron foam glass.

The statements so far show why during the production of the three foam glass types mentioned in the introduction to this dissertation (fine-celled foam glass, foam glass type 1, foam glass type 2) rigorous efforts were made to ensure that within each type (e.g. foam glass type 1) only one production parameter was varied to achieve different densities ρ.