Filler-hydrates Adhesion Properties in Cement Paste System: Development of Sustainable Building Materials

Fillers, such as limestone or quartz powder, are used as a replacement of Portland cement. Their application can make concrete more environment friendly and possibly cheaper. Additions of limestone or quartz powder have been reported to exert a limited chemical effect on cement hydration. The main quasi-chemical effect of added limestone and quartz powder is that they accelerate cement hydration by facilitating nucleation and growth of reaction products at their surfaces. Fine fillers in cement paste can result in improvements in strength because of a lower porosity and a denser packing. At the same time, however, the use of fillers results in dilution of Portland cement particles and their strength-providing reaction products in the paste. This ‘dilution’ effect will lead to an increased porosity. Above a critical amount of fillers, the effect of dilution exceeds the effect of packing, resulting in a lower strength of the hardened paste or concrete. These effects (porosity, packing and dilution) on the strength of cement paste have been studied intensively, but they (porosity, packing and dilution) cannot explain difference in performance of different fillers. Also adhesion between filler particles and reaction products has an influence on the strength of blended cement paste. However, filler-hydrates adhesion properties and their effect on the strength of blended cement paste are not quite well understood yet. The basic questions why filler particles and reaction products adhere to each other in blended cement paste, and how the chemistry and surface characteristics of fillers affect this adhesion, are rarely addressed and need further research. The aim of this project is to study the filler-hydrates adhesion properties in blended cement paste system. Firstly, the influence of the filler-hydrates adhesion properties on the strength of blended cement paste has been analysed. The compressive strength of cement paste blended with limestone powder and micronized sand was studied experimentally. The contact area between different solid phases in these cement pastes was quantified numerically. The relationship between the measured compressive strength and simulated contact area was analysed (“contact area concept”). Based on this relationship, the influence of the filler-hydrates adhesion properties on the strength of blended cement paste was quantified. It was found that micronized sand-hydrates contact area had no contribution to the compressive strength. By contrast, the limestone-hydrates contact area in cement paste had a substantial contribution to the compressive strength. Secondly, the filler-hydrates adhesion properties were studied at the microscale. Crack paths and fracture surfaces of loaded cement pastes were investigated by scanning electron microscopy (SEM) observation. Parallel with the SEM observations, the influence of interface’s mechanical properties on crack propagation, tensile strength and fracture energy was studied numerically by using a lattice model. Based on these SEM observations and simulation results, the mechanical properties of the interface between filler particles and hydration products were evaluated. Meanwhile, this provided a validation of the ‘contact area concept’. Then, the filler-hydrates adhesion mechanisms in blended cement paste system were investigated. The influence of the chemical nature of fillers on the interaction between main ions, i.e., Ca2+, SO42-, in the pore solution of blended cement paste and filler surfaces was investigated via zeta potential measurements. Meanwhile, microscopic observations of the nucleation and growth of C-S-H on the surface of these filler particles were performed by SEM. It was concluded that Ca2+ ions chemically adsorbed at limestone surfaces led to the formation of a relatively strong bond (most likely, ‘ionic-covalent’ bond) between a limestone particle and C-S-H. By contrast, Ca2+ ions electrostatically adsorbed at micronized sand surfaces resulted in an attractive ion-ion correlation force and hence a relatively weak bond between a micronized sand particle and C-S-H. This information about the filler-hydrates adhesion mechanisms is very important for the search for new fillers and for improving the performance of existing fillers. For example, based on the knowledge acquired in this study, carbonation can improve the performance of hardened cement paste powder when it is used as a filler in cement paste. This is because carbonation can turn the silicate and CH phase of the surface of the recycled hardened cement paste powder into calcite phase. This can enhance the adhesion properties between hardened cement paste particle and new hydration products. Finally, the fracture behaviour of cement paste with strong and weak filler-matrix interfaces was simulated at microscale by using a lattice model. The simulations indicated that the bond strength between filler particles and C-S-H matrix plays a more important role in the crack propagation and the strength of blended cement paste compared to the role of the particle size distribution, size (5, 10, 15 and 20 µm), shape, surface roughness and volume fraction (5, 15, 25, 35 and 45%) of the filler. These findings provided support to the previous findings that the strong interfaces between limestone particles and hydration products are due to the superior bond between limestone and hydration products rather than the physical surface properties, such as shape and surface roughness. Moreover, this study indicated the direction for optimization of the performance of fillers in cement paste in view of strength. To improve the performance of fillers, priority should be given to improving the bond strength between filler particles and hydration products. Modifying microstructural features (i.e., shape, surface roughness) of filler is less effective.