According to a previous study, it has been proven that the volume expansion associated with ice formation which causes cracking at a micronic scale, is not able alone on its own to induce the characteristic flaws observed on outdoor tiles. Therefore, additional stresses associated with laying are necessarily involved in the damaging mechanism. Based on this assumption, three tests taking into account the stresses associated with laying have been proposed and their reliability assessed. Two of them consist in applying successive frost-thawing cycles to structures composed of tiles either embedded into a mortar structure or clamped into a metallic frame. Another test based on submitting tiles having previously undergone frost-thawing cycles, to static compression has also been proposed as it allows a rapid appraisal of the frost resistance of those tiles. The three proposed tests enable at various degree to reproduce on tiles having the lowest frost resistance the characteristic flaws (spalling) observed on outdoor terraces. However taking into account practical considerations (assembly easy to carry out, no delay before testing) as well as its discriminating aspect, the test based on the use of a metallic frame appears as the most appropriate for further standardisation.
