Next: Use of lightweight Up: Modifications of ITZ Previous: Modifications of ITZ

Use of mineral admixtures

The modification of ITZ microstructure due to the addition of silica fume has been well documented by Scrivener et al. [14]. When properly prepared, concrete containing silica fume exhibits ITZ's that are nearly as dense as the bulk paste. This is due to both the small size of silica fume particles, which allows them to pack more closely near the aggregate surface, and their high pozzolanic activity, which converts CH into more CSH.

The simulations described in detail by Bentz and Garboczi [9] allow for both the size and reactivity of mineral admixtures such as silica fume and fly ash to be varied. Since in the 3D version of the model, a pixel is typically 1 micrometer on a side, well dispersed silica fume can be represented by small particles one pixel in size. Conversely, fly ash or applomerated silica fume is "large", and is assigned the same size as cement particles. Pozzolanic activity is varied by specifying how many units of CH each unit of mineral admixture may consume via the pozzolanic reaction. According to the reaction given earlier, one volume unit of pure silica would be able to consume 2.08 volume units of CH. Fly ash must be assigned a lower reactivity as it is typically only about 50% silica. Thus 1.04 might be a reasonable maximum activity to be expected for fly ash, while 0.47 appears to be a more likely value to be achieved in practice based on data from Hooton [15].

Since increased strength has provided a major impetus for the use of silica fume in concrete, results will be summarized in terms of the total phase fraction of cement + CSH + mineral admixture present, assuming these to be the major load bearing phases in the cement paste matrix. Figure 4 shows the distribution of these phase fractions near an aggregate for a variety of concretes at a constant water-to-solids weight ratio (w/s) of 0.45 after 77% hydration. In Fig. 4, size refers to the size of the admixture particles, while reactivity is equivalent to the pozzolanic activity discussed above. Small silica fume paricles added at a 20% replacement for cement, on a mass basis, result in a nearly uniform distribution of these load bearing phases throughout the microstructure. Fly ash is not nearly as effective as silica fume, although the results suggest that utilizing very fine fly ash should be beneficial if its ultimate pozzolanic activity of 1.04 is achieved due to the exposure of all available silica surfaces. Berry et al. [16] have obtained experimental results in support of this hypothesis. Thus small-sized pozzolanic mineral admixtures are seen to somewhat offset the wall effect while not affecting the one-sided growth effect, but they offer the additional benefit of converting CH into a larger volume of pozzolanic CSH.

Figure 4: Total load bearing phases vs. distance from aggregate for concretes containing 20% of various mineral admixtures.