Nonat – Early Hydration of Calcium Silicates: Driving Forces and Limiting Steps
André Nonat
Institute Carnot de Bourgongne, Universite de Bourgongne, France
Despite of a considerable number of papers, there is not a general agreement on the mechanisms and the control of the hydration of calcium silicates. In our mind it is mainly due to the fact that hydration is mostly studied in the context of cement hydration only, that is to say in a highly concentrated suspension (water to cement ratio close to 0.5); the studies are generally focused on the transformation of the solid, neglecting the aqueous solution even if a dissolution-precipitation process is now generally agreed. The peculiar shape of the heat evolution rate curve recorded during the early hydration of a tricalcium silicate paste is also probably at the origin of most of the misunderstandings.
To avoid useless hypotheses, it is convenient to address the different processes involved in C3S and C2S hydration on the light of the basic physico-chemical concepts. From the more general point of view, calcium silicate hydration brings into play four different phases, three solid phases and one liquid phase: one solid phase, C3S or C2S, is dissolving giving ions in solution in the liquid phase, and two are precipitating from these ions, C-S-H and calcium hydroxide, CH. The driving forces for both dissolution and precipitation are the respective departures from equilibrium (under and supersaturation respectively). The fact that the ions are the same in both processes implies the higher the undersaturation with respect to the anhydrous phase, the smaller the supersaturation with respect to the hydrate and vice versa.
From the kinetics point of view, the rate of each process is on the form R(t)=rinterfaceS(t) where rinterface is the interfacial rate of the process which depends on the super or undersaturation and S(t) is the extent of the surface of dissolution or precipitation. During hydration, all the silicate ions released by dissolution are consumed by the precipitation of C-S-H, so at each time, the following equation is satisfied:
R(t)=rdissinterface(bdiss(t))Sdiss(t)=rprecinterface(bprec(t))Sprec(t)
From this equation it is obvious that hydration may be controlled by:
– “geometrical parameters”, the extent of the dissolution surface or/and the precipitation surface
– “chemical parameters” controlling the interfacial rates.
The development of the C-S-H surface has been measured by proton NMR relaxometry during the first stages of hydration. It is found that the surface in contact with free water increases during the acceleretory period and passes through a maximum corresponding to the maximum of the heat release. During this time, the surface of C3S should continuously decrease. This is the growth of C-S-H which controls the rate of hydration during this period, the concentrations of ions in solution are then adjusted in such a way that both interfacial rates allows to satisfy the above equation.
Concerning the very early beginning, systematic study of nucleation of C-S-H proved that there is no induction period during hydration of C3S and that the amount of early hydrate is not sufficient to cover the surface of C3S. Moreover, an increase of these very early hydrates does not stop the hydration, on the contrary it accelerates.
Experimental studies of pure dissolution of C3S (very high W/C ratios) reveal a strong decrease of the dissolution rate with the decrease of undersaturation which explains by itself the first peak of the classical heat evolution rate curve.