Cross-section of concrete cracks caused by alkali-reactive chert and strained quartz. Filling the cracks and air voids is white alkali-silica gel.
B. Erlin Cross-section of concrete cracks caused by alkali-reactive chert and strained quartz. Filling the cracks and air voids is white alkali-silica gel.

Alkali-silica reactivity, better known as ASR, has openly been with us since the early1940s when it was discovered in California and publicized by T.E. Stanton. Over seven decades later, occurrences of this reaction may actually be increasing, despite our vast knowledge about ASR gained over these years. To explain why, we need to delve a little deeper into cement chemistry.

The alkalies of concern, primarily sodium and potassium hydroxides, Na(OH) and K(OH), (discussed in our columns and available at exist in variable but small amounts, primarily as alkali sulfates in all portland cements. They form alkali hydroxides as soon as water is added. These alkalies raise the pH of fresh concrete above 13, maybe even above 14, primarily because they are very water-soluble.

Now, quartz (SiO2), strained quartz, and their microcrystalline and non-crystalline relatives (cristobalite, tridymite, chalcedony, chert, opal, and acidic volcanic glass), are perfectly happy at pH of 12-12.5, the typical pH given in the literature for concrete. But they dissolve at higher pHs, producing alkali-silica gel. The gel is more voluminous than the siliceous components from which it formed (it occupies more solid space), is mobile, and under some conditions, causes localized stresses resulting in expansion and cracking.

Fortunately, pure, well-crystallized quartz (a frequent aggregate component) is dense and has such a small surface-to-volume ratio, that it generally is innocuous, meaning that it doesn’t react at normal ambient temperatures. Some researchers say, just wait, because given enough time, it will. The usual bad actors, though, are those varieties of silica having relatively large surface areas, like opal.

Reducing ASR

The measures usually taken to lessen ASR problems are to use low-alkali portland cement or to incorporate pozzolan into concrete mixes. When pozzolans are used to replace cement, the amount of cement in the concrete mix is lowered. This reduces the total amount of alkalies because of that cement substitution, although the silica of the pozzolans also rapidly ties up and immobilizes the alkalies. There are a number of ASTM tests used to identify alkali-silica reactive aggregates, and the effects of different alkali-bearing cements and pozzolans in accommodating the ASR reaction.

Nevertheless, in many areas of the country, the “good” aggregates have been used up, and “poorer” aggregates, of necessity or for unknown reasons, are being used. Because of that, and along with higher concrete cement contents, ASR occurrences have extended into areas and states previously reported to be free of ASR.

These occurrences have prompted new means to deal with the problem once it has outwardly manifested itself—usually in the form of cracking. The identity of the crack culprit and diagnosis of the problem usually begins with laboratory studies involving petrographic examinations that can be followed by length-change measurements of the concrete when exposed to a variety of laboratory conditions.

Its cure, sometimes accommodation, is difficult. Within the past decade or so, soluble lithium salts (for example, lithium hydroxide) have been developed that mitigate the ASR. A main difficulty is getting it into concrete within the concrete’s lifetime.

If the reactions have not destroyed the concrete’s strength, and the expansion is very slow, renewing closed joints by saw cutting to once again open them sometimes works. We recall a turbine pedestal inflicted with ASR that had not been diagnosed, where misalignment of the turbine, so vital for proper functioning of the high-spinning turbine blades, was a nightmare. The ASR expansion caused slow, minute expansion that progressively misaligned the turbine. Accommodation to the expansion was done by strategic, periodic, realignment of the anchors. But slow growth like that cannot be accommodated forever.

Where is ASR seen?

The fastest growing, outwardly manifested ASR reactions we have encountered resulted from particles in fine aggregate used in Minnesota, North Dakota, South Dakota, and Iowa. This was a glacier-deposited, easily mined sand containing shale particles. The shale particles contained opal-bearing marine microfossils. Because the shale particles had high permeability and the opaline microfossils had a large surface area, they rapidly reacted. Particles just below the surface reacted overnight to the extent that the next day stems of alkali-silica gel supporting up to 1-inch sized umbrellas of mortar grew from the finished concrete surface. Such “mushrooms” are evidence of how rapidly the reactions can advance under certain conditions. The solution in this case was to cure the concrete using limewater, which forms calcium alkali silica gel that has less or no expansive properties.

Alkali-silica gel is usually associated with delayed ettringite formation (DEF)—the school is still out on which came first, ASR or DEF. Maybe the chicken crossed the road to escape the encroaching ASR. That is the easy way out if you can find it. But where and how to cross the road isn’t an easy decision. So research is still ongoing on mechanisms of ASR, pre-emptive techniques, and once it’s there, accommodation with rehabilitation measures.

Knowing is not enough. Caring is better. Awareness helps provide the cure if it leads to elimination of the causes of ASR.

William G. Hime was a principal with Wiss, Janney, Elstner Associates and began working as a chemist at PCA 58 years ago. Bernard Erlin is president of The Erlin Co. (TEC), Latrobe, Pa., and has been involved with all aspects of concrete for 52 years. They wrote a regular column for CC for many years, collected into the book, The Concrete Intrigue, available through the World of Concrete Bookstore.