Investigation of Shrinkage Mechanisms in Hydrated and Carbonated Tricalcium Silicate Systems

238 p.Thesis (Ph.D.)--University of Illinois at Urbana-Champaign, 1983.Shrinkage-weight loss measurements, together with microstructural determinations, were obtained on fully hydrated tricalcium silicate (C(,3)S) paste that had original water/cement (w/c) ratios of 0.4 and 0.6. The surface area of the calcium silicate hydrate (C-S-H) phase determined by water is considered to be the single most important microstructural parameter determining the intrinsic shrinkage potential of paste hydrated C(,3)S. This is correlated to the Gibbs-Bangham mechanism, which relates shrinkage and swelling of high surface area components to changes in surface free energy of the solid phase caused by desorption or adsorption of water, respectively. It is active over the entire relative humidity range in which shrinkage occurs. At high humidities capillary forces result in additional shrinkage, which increases considerably with increase in capillary porosity. Irreversible shrinkage is thought to confirm the two shrinkage mechanisms proposed. The irreversible fraction of total drying shrinkage increases with time, exhibits a maximum at a relative humidity of about 0.50 and increases with w/c ratio of the hydrated C(,3)S paste. The shrinkage of C-S-H in C(,3)S pastes is not restrained by calcium hydroxide (CH) since the elastic moduli of CH and C-S-H are approximately the same. There is evidence that in carbonated unhydrated C(,3)S compacts C(')C blocks off some of the external surface area of the C-S-H thereby lowering the shrinkage potential.Helium pycnometry measurements on the solid phase of a 0.4 w/c fully-hydrated C(,3)S paste subjected to first drying at 0% RH show that the internal shrinkage of the C-S-H phase, including interlayer regions, is several times the external volume shrinkage. The solid density of C-S-H including CH and interlayer water at zero shrinkage was found to be 2.16 g/cm('3), and the density of externally adsorbed water calculated to be 1.23 g/cm('3). The removal of interlayer water apparently results in a nearly complete collapse of interlayer regions vacated by water. Pore volume measurements using water, methanol, helium, and nitrogen combined with mercury intrusion, are able to determine the original external pore volume. Only water can determine total pore volume including the interlayer spaces, although definition of total porosity is somewhat arbitrary.U of I OnlyRestricted to the U of I community idenfinitely during batch ingest of legacy ETD