Sweet answers to the decades-long debates on the sucrose retardation

Abstract

In the cement and concrete industry, the use of chemical admixtures is essential to improve either the properties for fresh pastes or the hardened concrete. However, many chemical admixtures cause cement hydration retardation in addition to their main effects. Among them, sugars are a family of retards which work specifically to extend the open time by delaying the onset of cement hydration. According to early studies, sucrose is a very effective retarder in the family of sugars. Many researchers have studied mechanisms of hydration retardation caused by sugars. Despite different hypothesis being proposed, the retardation mechanism of sucrose is still not fully understood. This can be attributed to the complexity of cement system which contains different phases, and the coupled chemical reactions taking place during each hydration period. In presence of sugars, the interaction between these molecules with different mineral surfaces as well as their behaviors at the solid-liquid interfaces adds more difficulties to such research. In this thesis, synthetic tricalcium silicate was used as the first step to simplify cementitious systems. Due to its good ability in delaying hydration, sucrose was chosen as an example of chemical admixture for studying the mechanism of hydration retardation. Isothermal calorimetry measurements were performed in situ to follow hydration processes in a more reliable way. Different models were proposed for estimating sucrose adsorption from its dosage, which gave the possibility to build up the relation between rate of hydration in the induction period and sucrose adsorption, the latter being best described in terms of surface coverage. It was shown that the hydration velocity of tricalcium silicate in the induction period decreases linearly with surface coverage of sucrose. Better said, that rate is proportional to the fraction of surface not covered by sucrose. In case where portlandite was mixed with tricalcium silicate, the hydration rate in the induction period of this blended material is enhanced by the increase surface area, which we interpret as allowing faster deposition of hydrates. Sucrose however does not seem to modify this, or to be displaced from tricalcium silicate to portlandite. With understanding on hydration kinetics for tricalcium silicate, the second step in this work was to apply the same experimental and analytical methods to an industrial cement. It was found that the temperature dependence of sucrose adsorption behaves differently on cement as on tricalcium silicate. As to the impact of sucrose on the rate of hydration in induction period, it was interesting to notice that in the low dosage range, it only depends on the surface coverage of sucrose on tricalcium silicate. Regarding the hydration of cement in presence of portlandite and sucrose, it seems that sucrose preferably adsorbs on the surface of tricalcium silicate phase, the transfer of sucrose molecules from this surface to portlandite surface does not take place. In addition to these works, we also wanted to achieve direct “observations” for sucrose behaviors during hydration with NMR measurements, because of their selective and high sensitivity on NMR active isotopes. For detecting the sucrose in pore solution, 1H refocused INEPT NMR was successfully developed with largely enhanced sensitivity together with a quantification method. This opens the possibility of detecting mobile sucrose from a paste. However, monitoring sucrose adsorption during hydration of tricalcium silicate with this method posed many problems. Observations of sucrose in hydrated tricalcium silicate system were successful. But we failed in seeing the interaction of sucrose with this mineral surface, even though 2D HETCORE NMR and some pulse sequences based on different covalent pulse transfer were tried. For this question, maybe other surface enhanced NMR techniques can be tried in the future, such as dynamic nuclear polarization (DNP) may be more appropriate.