Adsorption de polycarboxylates et de lignosulfonates sur poudre modèle et ciments

Placing concrete requires much more water than the cement needs for its hydration. This results in an important porosity in the hardened concrete, which accentuates the degradation of this material. By adding small amounts of polymeric admixtures, called superplasticizers, to the fresh concrete one can significantly reduce the amount of water required to obtain the suitable workability. The plasticizing effect of superplasticizers has been studied for many years, and remains today, an ongoing field of research. This research led to a better understanding of the forces that act during the deflocculation and dispersion of cement grains induced by superplasticizers. This research work was performed within the framework of the "Superplast" European project. Its main objective has been to determine which parameters (molecular structure, induced charge, adsorption mode, molar mass) influence the plasticizing effects of superplasticizers. The results of this study will allow synthesis new polymeric admixtures with better performances than those currently available. For this study, two types of superplasticizers were selected: lignosulfonates and polycarboxylates. For each one, different samples were synthesised and characterised. For each sample, adsorption, rheological and interaction forces measurements were performed by our partners while we have performed adsorption and electroacoustic measurements on model powder suspensions. Different cement model powders were investigated and a magnesium oxide was selected. The particle size distribution and reactivity were carefully characterized. The inert model system (MgO) allowed us to study adsorption mechanisms without the complexity linked to cement hydration reactions that modify surface and solution of the suspension. Particle surface charge and pH are the two main parameters that influence the polymer adsorption and the polymer conformation. Magnesium oxide which has a high isoelectric point (around at pH 12.4) allows a surface charge similar to cement suspensions at high pH. First, we have measured the adsorption isotherms of all superplasticizers on model suspensions in NaOH (0.01M). This study allowed us to evaluate the affinity of each polymer for MgO and its adsorption plateau. These measurements showed that the lignosulfonates have a higher affinity than the polycarboxylates. They also showed that lignosulfonate adsorption is mainly influenced by their molar mass and their carboxylic group content (for similar sulfonate group content) while polycarboxylate adsorption is mainly driven by the backbone length, the side chain length and the carboxylic group content. Adsorption plateaux allowed us to calculate the surface coverage ratio and to estimate the superplasticizer conformation on the surface. Adsorption isotherms were finally measured on a Portland cement. The polymer adsorption is influenced by the same parameters as on MgO. The model system MgO is representative for the polymer adsorption on cement. In a second step, different electrolytes were added to model suspensions. This practice allowed us to study separately the effect of the main ions present in cement suspensions (Na+, Ca2+, SO42-, OH-) and to mimic the ionic composition of the aqueous phase of the cement suspension. Lignosulfonate adsorption is neither influenced by the studied ions or the pH. Only an increase of ionic strength increases the adsorbed polymer mass. Polycarboxylate adsorption is influenced by calcium and sulfate ions and by the particle surface charge. Finally, superplasticizer adsorption was studied on different cements by our partners and on a fly ash and a silica fume by ourselves. Lignosulfonate adsorption isotherms on cements showed that affinity and adsorbed polymer mass increases with the C3A content and decreases with the alkalis content. The isotherms measured on industrial by-products showed that superplasticizers adsorb on fly ash, but not on silica fume.