Application of the Rietveld-PONKCS technique for quantitative analysis of cements and pitfalls of hydration stopping methods

Existing methods to quantify the degree of hydration of cementitious materials such as selective dissolution and image analysis of scanning electron micrographs are either laborious or unreliable. Meanwhile, quantitative X-ray powder diffraction (QXRD), routinely used to study kinetics and phase evolution of hydrating cements present opportunities to determine quantities of crystalline and poorly-crystalline phases simultaneously. The profile fitting technique however requires structure files, which are non-existent for poorly crystalline materials including most supplementary cementitious materials (SCMs). This contribution is focused upon developing a pseudo-structure file for ground granulated blast furnace slag (GGBS), a Phase of No Known Crystal Structure (PONKCS) for implementation in the Rietveld refinement. Factors affecting the developed model and its stability are assessed. Following, the model is used to quantify the residual GGBS content in hydrated composite cement. The effect of hydration stoppage technique on accuracy of the PONKCS phase is assessed on a binary slag cement. The results show that, the PONKCS phase was stable in synthetic and hydrated cements. Hydration stopping methods that modified the background through decomposition of phase assemblages eg freeze-drying also caused overestimation of the PONKCS phase. The QXRD/PONKCS technique is less laborious, has good consistency with the quantified crystalline phase, and enables the degree of hydration of SCMs to be measured alongside hydrated phase assemblages.