Properties of concrete - Concrete Mix Design, Quality Control and Specification - page 89

as well as structures exposed to periodic wetting. Carbonation certainly
makes potential chloride-induced corrosion worse by releasing bound
chloride as well as reducing alkalinity.
Because of the sensitivity to the moisture content, the possible difference
in carbonation rates between a lab test and site exposure can be very large.
Where practical if there is a dispute over the potential carbonation, the
authors would suggest periodic sampling to measure in situ carbonation
over the coming few years. If in situ rates show that there is potential for
carbonation through cover within the proposed design life, the concrete
can be coated later. Carbonation is a slow process and therefore the time
to intervene to prevent the carbonation front reaching the reinforcement
is long.
5.1.2 Alkali-aggregate reaction
An important cause of deterioration in concrete is alkali-aggregate
reaction, which can include both alkali-carbonate reaction (ACR) and
alkali-silica reaction (ASR). Reactive carbonate rocks are relatively rare;
if they are suspected, potential ACR should be established by testing to
ASTM C586, and if present, they should be avoided. Alkali-silica reaction
is a disruptive expansion of the cement matrix arising from the combi-
nation of alkalies (usually, but not necessarily solely, from the cement)
and reactive silica within the aggregate. Although generally confined to
particular geographies, the phenomenon can be disastrous when it does
occur. There are three possible strategies to limit its occurrence. One is to
limit the quantity of total alkalies (sodium and potassium) in the cement
to less than 0.6% calculated as Na 2 O equivalent (1 × Na 2 O + 0.685 K 2 O).
Another is to test the aggregate for reactivity. A third possibility is to
provide an excess of reactive silica in the form of fly ash, silica fume, or
natural pozzolan so as to consume excess alkali present in a nonexpansive
surface reaction product. Iceland, which had widespread problems with
ASR due to reactive aggregates and a high alkali cement, has virtually
eliminated ASR since 1979 with the introduction of 7.5% silica fume into
its cement as well as washing of sea dredged materials and limiting the
use of reactive materials. Queensland in Australia has also largely solved
its ASR issues by requiring a minimum fly ash replacement of 20% of the
component Portland cement.
5.1.3 Sulfate attack
Sulfate attack is an important deterioration mechanism of the concrete
itself. Sulfates react with the portlandite and tricalcium aluminate in
hydrated concrete to cause disruptive expansion. Sulfate resisting (Type V)
cement has a limited tricalcium aluminate content. Low heat Portland