No abstract

The validity of the statistical nanoindentation (SNI) technique for cementitious materials has been questioned by numerous researchers; however, it is still a commonly used method in literature. This study revisits assumptions pertaining to continuum mechanics of the method by using a well-established selective SNI method to evaluate continuity in the mechanical response of each indent. Ordinary portland cement (OPC) was investigated alongside a two-phase aluminosilicate ceramic (Al-Si). Filtering for continuity of indents eliminated a majority of indents in each sample. The indents meeting continuity requirements in the OPC sample had moduli values in the range of documented calcium silicate hydrates and portlandite-the predominant hydration phases in OPC. In the Al-Si sample, 73% of continuous indents were assigned to a mullite phase, although the chemical analysis showed only 42% mullite. This was attributed to the embedded morphology of the mullite phase, which made it more susceptible to structural compliance. These observations suggest that SNI may be ineffective for measuring the mechanical properties of individual phases in a microstructure if they are not prolific enough to exhibit a continuous indent response. Discontinuous indents due to microstructural interaction between phases were also shown to have a potential influence on the number of clusters (i.e., phases) suggested by the deconvolution algorithms.