Durability of concrete and cement composites

Physical and chemical characteristics of cement composites

The aim of this chapter is to provide information on aspects of the internal characteristics of concrete that relate to concrete durability. Pertinent chemical and physical characteristics, particularly those involving pore solutions and pore structures, are discussed in some detail, as these most closely affect most durability concerns.

Concrete is an unusual engineering material. Unlike most engineering materials it is held together by a porous binder, specifically a complex of solids and pores called `hydrated cement paste’. The binder is the continuous phase in the composite; thus the fact that it is porous is critical with respect to movement of water and chemical substances into or out of the concrete. This is a feature extremely relevant to the durability of the concrete in service.

A further characteristic that sets concrete apart from most other engineering materials is that the porous binder is intrinsically hydrous ± that is, except for any residual unhydrated cement, it is made up of compounds that are all hydrated solids. These compounds (calcium silicate hydrate or C-S-H, calcium hydroxide, ettringite, monosulfate, and others) develop spontaneously within the structure as the result of chemical reactions between water and ground Portland cement. They are deposited within the particular internal chemical environment of the specific concrete, and in service they are in at least temporary equilibrium with that internal environment. It is sometimes not fully appreciated that these hydrated compounds are subject to reorganization if the local internal chemical environment is modified ± by leaching, by intrusion of dissolved salts, or by other processes. This is the unfortunate origin of many durability problems.

Yet another distinction sets concrete apart from most other engineering materials. Not only is the hydrous binder porous, but at least some of the pores contain a highly concentrated solution of alkali hydroxide. Under some unusual exposure conditions, the pores within a given concrete structure may be fully saturated with this solution. Generally they are not; the larger pores may com- monly be empty, especially those near concrete surfaces that have been exposed to evaporation, or within concretes that have been mixed at low enough water: cement (w:c) ratios that they are subject to self-desiccation. On the other hand, even prolonged exposure to dry conditions in service does not fully empty the finest pores. The actual concentrations of dissolved substances in the solution retained in the pores of concretes may be significantly affected by leaching, by partial drying, or by intrusion of ions and other dissolved substances from the outside of the concrete. The details of the solid pore structures bordering and defining the pores of the binder in a given concrete play a significant role in most concrete durability problems. Since the aggregates used in most concretes have few interconnected pores, the paste pore structures almost always govern the rates at which water or ion movements can occur. In this connection it is often suggested that permeability to water is the single most important pore structure-related characteristic controlling the potential durability of a given concrete. Strictly speaking, permeability refers to rate of mass transfer of a fluid (usually water) as a function of applied hydraulic head. This actually may be less important in the context of concrete durability than other related parameters, such as the rates of ion transfer, or rates of internal water vapor transfer in unsaturated concretes. Because of these various concerns, the present writer proposes to use a more general term `permeation capacity’ instead of `permeability’ in discussing the durability-related transport characteristics of concretes. It is noted that a dictionary definition of the verb `to permeate’ is `to penetrate through the pores, interstices, etc.’, without any particular permeating substance or mechanism being specified. It is well known that the permeability of concrete is a strong function of its water:cement (w:c) ratio, and within concrete of a particular w:c ratio, of the degree of cement hydration. Figure 2.1, the classical relationship derived from Powers et al.,1 illustrates the extremely non-linear effect of the w:c ratio on the permeability, in this case as measured in thin specimens of almost fully hydrated (93% hydrated) cement pastes. A similar trend in the effect of w:c ratio on permeability is usually expected with concretes. It is well understood that permeability of young concretes is initially high and decreases with degree of hydration. The degree of hydration experienced in most field concretes is limited, especially in those of low w:c ratios; practically speaking, concretes almost never actually approach the degree of hydration of the paste specimens measured by Powers et al.1 Other characteristics of concrete that measure its permeation capacity, such as ion diffusion coefficients or electrical conductivity values, generally reflect tendencies similar to those of permeability with respect to variations in w:c and degree of hydration. However, some of these measurements may be influenced by factors that do not necessarily affect permeability measurements. For example, the chemistry of the particular cement used exerts a major influence on the concentration of ions in the pore solution of the concrete, and therefore on its electrical conductivity. The incorporation of silica fume, fly ash, or other supplementary cementitious components in concretes exerts a very strong influence in reducing permeation capacity, although not necessarily to the same extent in different types of measurements. Similarly, the conditions under which hydration takes place, especially whether or not the concrete was steam cured, may exert important effects. Once placed in a structure, concrete is exposed to in-service conditions that may alter both pore solution composition and pore structures. Entry of chemical substances, leaching, repeated wetting and drying, freezing events, carbonation, etc., all may induce significant internal changes in pore fluid composition and in pore structure. Some of these alterations may induce specific durability problems.

Often durability problems such as alkali silica reaction (ASR), steel
corrosion, delayed ettringite formation (DEF), etc., result in expansion-induced cracking. Cracking may also result from shrinkage rather than from expansion. However induced, cracks inevitably increase the effective permeation capacity compared to that of similar intact concrete. Attempts to predict potential service life of concrete structures often founder on how to account for the effects of future crack development on the expected permeation capacity. The present chapter is designed to provide an overview of internal chemical and physical characteristics of cement paste binders in concretes, especially focused on pore solutions and pore structures. Some discussion of permeation capacity-related measurements that depend on interconnections between pores and in certain cases on the effects of pore solutions is also provided. In this chapter the writer has deliberately chosen to confine himself to findings based on experimental evidence, and has omitted mention of a parallel literature in which the microstructures and processes considered here are modeled, rather than investigated.


Related Articles