Q.: Reports on chloride corrosion of steel are confusing to me. Sometimes I get the impression that it is caused by deicers and other times it sounds as though calcium chloride admixtures are the culprit. Or can it be either one or both?
A.: Anything that introduces chloride ion into concrete--whether an admixture, a deicer, chloride-containing mix water, or sand washed with sea water--may cause the steel to corrode. Some authorities believe any chloride that gets into concrete will cause steel to corrode, given sufficient time. The question of whether chloride attack can be prevented or only delayed is complicated because there are several factors that slow down the rate of corrosion. If there is no corrosion after 50 years does this mean there will never be corrosion, or that there will be corrosion only if conditions change, or that there will eventually be corrosion no matter what? Some factors that prevent chloride corrosion or minimize it are concrete cover of 1 1/2 inch or more, permanently dry condition of concrete, low chloride concentration, or chloride present only in a nonsoluble form (as in chemical combination with some constituents of portland cement).
Although there has long been a recommendation that calcium chloride never be added to concrete mixtures in amounts higher than 2 percent by weight of the portland cement in the mix, stricter limitations have recently been proposed. See "Guide to Durable Concrete," by ACI Committee 201, Journal of the American Concrete Institute, December 1977, pages 573 to 609, especially Section 4.5.4 on page 593.