Next: Research Significance
Up: Main
Previous: Main
At the macroscopic level, concrete is a composite material consisting of discrete aggregates (sand, rock) dispersed in a continuous cement paste matrix. As with other composites [1], the bond between these two major components of concrete is a critical component of mechanical performance. The bonding region or transition zone in concrete between matrix and aggregate has therefore been a subject of considerable research interest in recent years. Both qualitative and quantitative characterizations of the interfacial region have been conducted [2,3,4,5]. The general consensus reached by these studies is that the transition zone is a region up to 50 micrometers wide containing a microstructure that is quite different from that of the bulk cement paste. The microstructure in the transition zone is characterized by high porosity, composed of relatively large pores, and often contains large crystals of calcium hydroxide. This transition zone is not unique to aggregate surfaces but is also present at other interfaces in cement-based composites such as the steel-cement paste interfacial region [6]. The existence of this interfacial region has been conjectured to be due in part to particle packing effects [5]. When particles are packed in a container, there is always a region near the container's edge where the porosity is higher than in the bulk. The width of this region is of the order of the largest particle's diameter, and is attributed to the presence of an edge preventing the most efficient (random) arrangement of particles. The experimentally-measured size of the interfacial zone, ranging up to 50 micrometers, is consistent with the particle-size distribution of portland cement, since portland cement particles average 15-20 micrometers in diameter, and particles exist that are up to several times larger. Recent two-dimensional simulations of the interfacial zone in concrete by the authors have verified this packing effect and identified a secondary effect, termed the "one- sided growth" effect [7]. In the bulk of the paste, hydration products are growing into any given area from all directions, but very near an aggregate, growth is occurring from only one side, the cement paste matrix side, so that volume in this region is filled less efficiently. This effect is of shorter range than the particle-packing effect. In certain cases, other factors may also contribute to formation of the transition zone, such as increased bleeding of the cement paste near an aggregate surface [8]. In the latter half of the 1980s, so-called high performance concretes have been developed. By utilizing a low water-to-cement (w/c) ratio cement paste containing both silica fume and a superplasticizer, along with high strength aggregates, high strength concretes, roughly defined by having compressive strength greater than 100 MPa, can be produced. In addition to strength, other properties such as durability and permeability may also be improved in these new materials [9,10]. One of the major effects of the addition of silica fume to concrete is modification of the transition zone. Generally, the presence of silica fume results in a denser, less porous, and more homogeneous microstructure in the interfacial region compared to ordinary portland-cement concrete [11,12].
This densification of microstructure, along with the observation that silica fume-containing concrete may actually be stronger than its component paste, has prompted Bentur et. al. to postulate that strength enhancement is due to a change in the role of the aggregate in the concrete composite [13]. That is, in ordinary portland-cement concrete, in terms of strength, the aggregate functions basically as an inert filler due to the presence of the weak interfacial zone, so that the composite concrete is weaker than the plain cement paste [13]. In concrete containing silica fume, however, this weak link is eliminated so that the aggregates can effectively act as reinforcing inclusions. Thus, the concrete containing silica fume may actually be stronger than cement paste containing silica fume.
As a mineral admixture, silica fume possesses two positive attributes. First, it is a very fine material, with an average particle size of 0.2-0.4 micrometers [14], so that the initial high porosity in the interfacial zone due to silica fume packing alone only extends out from the aggregate edge a few micrometers at most. Of course, the initially higher porosity region due to the packing of the cement particles still exists, albeit modified by the silica fume packing. Second, silica fume is highly pozzolanic, so that it is able to react with the calcium hydroxide (CH) formed during hydration to produce "pozzolanic" (or secondary) calcium silicate hydrate (C-S-H) Conventional cement chemistry notation is used throughout this paper: C=CaO, S=SiO2, H=H2O, A=Al2O3, and F=Fe2O. As proposed by Goldman and Bentur [15], both of these attributes may contribute to the aggregate-paste bond improvement provided by silica fume. The fineness of the silica fume allows it to pack efficiently more closely to the aggregate surface, where subsequent pozzolanic hydration will result in the production of a denser, more homogeneous microstructure.
In this paper, the two-dimensional simulations of the interfacial zone developed in Ref. [7] are extended in two ways. First, the simulations are conducted in three dimensions to more closely correspond to real concrete. Second, a variety of mineral admixtures (inert, silica fume, and fly ash) have been incorporated into the model to assess their effects on the transition zone, and to critically examine the above ideas concerning transition zone properties.