From the foregoing sections, it will be evident that guidance on measures to minimise the risks of premature degradation of concrete from the different forms of sulfate attack has had to become increasingly complex in recent years. While it must allow for the varied effects of different exposure environments, it must also take account of the complex chemistry of interaction between sulfates and the constituent phases and microstructure of the hydrating cement. This can make for rather unwieldy guidance documents that are difficult to apply, even when they are restricted to a particular location. Thus in the UK, where a series of Building Research Establishment Digests has long formed the main basis for advice in British standards on the requirements of concrete exposed to sulfate- bearing soil and groundwater, the guidance published in Building Research Station Digest 90 (1968) had a simple one-page table in which exposure sites were divided into five classes on the basis of sulfate level in either soil or groundwater. Progressively tighter limits on the types of cement suitable and on cement contents and water to cement ratios were imposed for increasing severity of sulfate class. An essentially similar approach was followed by successive documents in the series until BRE Digest 363 was revised in 1996 (BRE, 1996) to draw attention to the fact that the risks of TSA required explicit recognition.
In marked contrast to the above, however, the 3rd edition of BRE Special Digest 1 (2005) is divided into six parts and runs to no fewer than 62 pages. In dealing with the chemical aggressiveness of the ground, it sets out procedures for determining the Design Sulfate Class (DS Class) from the soluble sulfate (and magnesium) and the potential sulfate (in cases where sulfide-bearing minerals may oxidise); it then proposes the classification of natural and brownfield sites in terms of an Aggressive Chemical Environment for Concrete Class (ACEC Class), dependent on the DS Class together with the pH and mobility of groundwater. Its recommendations for the specification of an appropriate quality of concrete, the Design Chemical Class (DC Class), to be used in contact with the ground are formulated from consideration of the ACEC Class together with the hydraulic gradient of the groundwater, the type and thickness of the concrete element, and the intended working life (50 or 100 years). Additional protective measures (APMs) are recommended for dealing with certain highly aggressive conditions, whilst relaxation of the recom- mendations for deriving DC Class are proposed for surface carbonated precast concrete (for reasons explained in Section 4.3) and for specific precast concrete products that are manufactured under conditions of rigorous quality control and considered to be of low permeability. Simple it is not. At the time of writing the present chapter, BS 8500 (2002) was being updated on the basis of guidance presented in BRE Special Digest 1, 3rd edition (2005).
A final comment that should be made here on the scope of BRE Special Digest 1 is that, while recognising that chloride ions are commonly found in soils and groundwaters in the UK, the document is restricted to dealing with situations where the levels of chloride concentration do not exceed those found in brackish water in brownfield sites (12000±17000 mg/l). In these cases, the chemical effects of chloride on concrete are considered to be innocuous. The effects of higher concentrations of chloride salts are obviously of relevance to the performance of concrete structures in marine environments and where de- icing salts are applied, but these situations are not dealt with in BRE Special Digest 1, which makes reference to BS 6349-1 (2000) for maritime structures and to BS 8500-1 (2002). It may be noted that, in such cases, the most important effects of chlorides are on the corrosion of reinforcing steel in concrete (considered in Ch. 5) and, in certain locations, the severity of freeze/ thaw damage (considered in Ch. 7). With regard to the performance of coastal and offshore structures, however, it is of interest to record that, although seawater is of variable composition depending on location, it often contains around 3.5% by weight of salts with a pH in the region of 7.5±8.4 and, while its main ionic components are Na+ and Clÿ, it also contains SO4 2ÿ at concentrations that would seemingly correspond to a DS-3 classification as well as significant levels of Mg2+ ± see Table 4.2. The chemical action of seawater on concrete is therefore due to the influence of several reactions proceeding concurrently and, although the main concerns in this area have long been ascribed to the presence of MgSO4 (Vicat, 1857), it has often been found that the effects of the SO4 2ÿ and Mg2+ in seawater are considerably milder than those that would be produced simply by exposure of concrete to solutions of pure MgSO4 of similar concentrations (Lea, 1970; Taylor, 1997). As seawater varies considerably in composition and temperature from one location to another, and so does the severity of numerous related physical factors (wave action, abrasion, freezing, etc.), it is obvious that generalisations about its degradative effects on concrete are apt to prove misleading. The subject has been extensively researched in many parts of the world and further information is available from a number of sources, e.g. (ACI, 1980; RILEM Technical Committee 32-RCA, 1985; Mehta, 1991; Hobbs and Matthews, 1998; Mehta, 1999; Thomas et al., 1999). As the ancient marine structures that were referred to in Ch. 1 and others of more recent origin illustrate, the production of mortars and concretes that are capable of resisting the chemical effects of seawater over many years is quite possible if steps are taken to achieve appropriately low penetrability. The most important requirements are that the material should be made at suitably low water/cement ratio with adequate cement content, well compacted and properly cured; the use of pozzolanic or slag cements has also been found to provide enhanced durability in some situations (Massazza, 1998; Moranville-Regourd, 1998).
