This work is the second part of an overall project, the aim of which is the development of general mix design rules for concrete containing different kinds of mineral admixtures. The first part presented the separation of the different physical effects responsible for changes in cement hydration when chemically inert quartz powders are used in mortars. This second part describes the development of an empirical model, based on semiadiabatic calorimetry measurements, which leads to the quantification of the enhancement of cement hydration due to the heterogeneous nucleation effect at short hydration times. Experimental results show that not all the admixture particles participate in the heterogeneous nucleation process. Consequently, the concept of efficient surface S-eff is introduced in the-model. S-eff is the total admixture surface S (m(2) of mineral admixture/kg of cement) weighted by a function xi(p). The efficiency function xi(p) depends only on the replacement rate p and is independent of time, fineness and type of mineral admixture used. It decreases from 1 to 0: Low replacement rates give an efficiency value near 1, which means that all admixture particles enhance the hydration process. An efficiency value near 0 is obtained for high replacement rates, which indicates that, from the hydration point of view, an excess of inert powder does not lead to an increase in the amount of hydrates compared with the reference mortar without mineral admixture. The empirical model, which is mainly related to the specific surface area of the admixtures, quantifies the variation of the degree of hydration induced by the use of inert mineral admixtures. One application of the model, coupled with Powers' law, is the prediction of the short-term compressive strength of mortars. (c) 2004 Elsevier Ltd. All rights reserved.
