Thaumasite can be formed below 15ëC (ideally at 0-5C) by two sets of reactions that are very slow to get going. These reactions are known as the direct route and the woodfordite route (Bensted, 2003a). Note: Woodfordite is the mineral name for the partial solid solution that can arise between ettringite and thaumasite.
In the direct route, there is a general reaction of sulfate with carbonate, silicate (from the main cementitious binder C-S-H) and excess water in the presence of calcium ions. The reaction is very slow and normally takes several months to obtain a significant yield. Both the alite and belite phases provide the binder, C-S-H, by normal hydration, which can then react with the other ingredients to form the non-binder, thaumasite. This reaction may be represented in a very simplified way as follows:
The woodfordite route is also slow, but somewhat quicker than the direct route. This situation arises because there is already an octahedral arrangement for Al (and Fe) in ettringite, into which Si from the C-S-H can displace Al (and Fe) initially by solid solution to form woodfordite and then to `overwhelm’ the Al (and Fe) by exsolving of the latter. The portlandite formed readily carbonates to give calcite Cc, which can serve as a reactant for producing more thaumasite.
Sulfate Resisting Portland cements, although discouraging ordinary sulfate attack, associated with expansive ettringite formation, are no better than Ordinary Portland cements when subjected to thaumasite sulfate attack because the main cement binder, C-S-H, is a reactant for producing the non-binder, thaumasite. As an illustration of this and a demonstration of the influence of temperature on thaumasite formation, the effects of exposing hydrated specimens of high w/c made from 50% SRPC/50% limestone powder to MgSO4 solution (18000 mg/l with respect to SO4) for 100 days at 5ëC and 20ëC are shown in Fig. 4.2 (Francis, 1999). Magnesium sulfate attack reinforces thaumasite sulfate attack and makes the overall deterioration worse. Both the direct and the woodfordite routes are involved, with thaumasite and brucite (MH) being formed as products (Bensted, 2003a). Thaumasite is not just a form of sulfate attack, but of carbonate attack, too. Some interesting studies have been made of the role of carbon dioxide in the formation of thaumasite (Collett et al., 2004). The authors have stated that relatively high pH values, generally > 10.5, are needed for thaumasite formation and these conditions are maintained by the continual dissolution of portlandite from the cement hydration products in the presence of aggressive CO2 and bicarbonate ions dissolved in groundwater, suggesting a requirement for portlandite to be present as a reactant. This, however, overlooks the fact that other cement hydrates, such as C-S-H, also buffer the pore solution at pH values > 10.5 so thaumasite can be expected to form without portlandite per se being present as an initial reactant. The influence of stereochemistry seems to play a crucial role in governing the reaction kinetics as it would appear that thaumasite is not produced until a stable transition state intermediate can form, having six OHÿ groups around a highly polarising Si4+ cation with nearby carbonate groups to assist in charge delocalisation away from the silicon cation, in order to give a stable thaumasite structure.
