Title: Investigation of Dispersion Levels of Silica Fume in Pastes, Mortars, and Concrete

Author(s): D. Baweja, T. Cao, and L. Bucea

Publication: Symposium Paper

Volume: 212

Issue:

Appears on pages(s): 1019-1034

Keywords: densified; dispersion; durability; high performance concrete; scanning electron microscopy; serviceability; silica fume; strength regression; undensified

DOI: 10.14359/12736

Date: 6/1/2003

Abstract:
A significant amount of work has been carried out on silica fume and its use in concrete, both from a research and field perspective, over the last 30 years. There is currently, however, a major debate on silica fume dispersion in concrete and the impact of undispersed silica fume on short and long-term concrete properties. Some suggest that complete dispersion of silica fume can be achieved in concrete through appropriate mixing strategies. Others disagree, suggesting that there are high probabilities of formation of silica fume agglomerates in concrete. The latter further suggest that late-reaction between silica fume agglomerates and the lime within the concrete matrix can po- tentially induce microcracking, and durability and serviceability problems in concrete elements. This paper presents results, obtained from an investigation into the use of silica fume in concrete. The work considered scanning electron microscope data on samples obtained in laboratory studies and in field concretes. Three silica fumes were investigated in pastes and mortars at a series of dose rates and used in conjunction with different superplasticisers. Investigations were also conducted on concretes from samples taken from actual structural elements. Typical 28-day compressive strengths for this concrete were 55 MPa. Examination of hardened silica fume pastes revealed a large number of agglom- erates. Increased mixing time did not significantly influence the presence of such agglomerates, although the use of superplasticisers appeared to marginally reduce agglomerate occurrence. In hardened mortars, silica fume agglomerates were again observed. Mixing time and superplasticiser dosage were found to influence agglomerate occurrence. In concrete, agglomerate occurrence was significantly reduced when compared with that found in the pastes and mortars. Although silica fume agglomerates were found in concrete samples, shear action of aggregate during mixing reduced silica fume agglomerate numbers significantly. No evidence could be found of damage to concrete elements resulting from the delayed hydration of undispersed silica fume.