ABSTRACT
The knowledge that some mineral aggregates are reactive with cement is not new. Indeed, this is the reason why many natural pozzolans can be activated by cement or lime. However, while many reactions occurring internally in cemented systems have beneficial consequences, or at least have no significant adverse effects, the alkali-aggregate reaction does give cause for concern. Its harmful effects arise from the expansive nature of the products of reaction, which lead to cracking. In turn, cracking may accelerate other mechanisms of deterioration, typically those likely to lead to loss of physical integrity, e.g. freeze-thaw, or those which arise from ingress of foreign ions: sulphate ions from saline waters, salts from de-icing run-off, etc. If, however, the alkali-aggregate reaction results in undesirable physical consequences, why study its chemistry? The answer to this question is, in general terms, straightforward: the physical process of expansion is generated as a consequence of chemical reactions. This potential for reaction is conditioned by the chemicals and mineralogical nature of the components of the concrete system-cement, aggregate, water, etc.—as well as by its conditions of service-temperature, humidity, etc. The magnitude of the expansive potential is, however, complex and as yet poorly understood, arising from a mixture of chemical and physical causes: factors such as the microstructure of the paste and the sizing and distribution of the reactive aggregate influence the magnitude of the observed expansion. It is important, therefore, to recall that the chemical and physical components of the reactions are virtually inseparable. However, the basic origins of the chemical potential involved in alkali-aggregate reactions is also clear: certain siliceous materials which occur in natural mineral aggregates react with the alkaline components of cement, and the increased molar volume of the products, relative to those of the reactants, creates swelling pressures. When these pressures cannot readily be accommodated or restrained within hardened concrete, it expands with cracking. The kinetics of the alkali-silica
reaction are not well understood, and quantitative characterisation of the potential for further, continuing, reaction at any particular point in time has not been achieved. As a consequence it is not yet possible to measure a single chemical parameter, or even several parameters, and correlate these with the expansive potential of the system. Thus it is not at present possible to relate chemical factors quantitatively to the prime engineering parameterdimensional stability. But an understanding of the basic chemistry has been achieved.
