By Wendell Dubberke
The alkali carbonate reaction (ACR) relates to faulty calcite and dolomite aggregates that perform poorly when placed in the alkaline environment of portland cement concrete (PCC).
The use of ACR to describe a certain type of PCC deterioration should be discontinued because it is too general. Its like going to a doctor, after not feeling well, and after a two hour checkup, the doctor tells you that you are sick. There are a variety of sicknesses and you would like a little more detail. Like illnesses, there are a variely of reasons why certain carbonate aggregates do not perform adequately in PCC.
For some carbonate aggregates, the difference between performance and non-performance, in PCC, depends on whether deicing salts were applied to the surface or if they were exposed to brine from other sources.
In some cases, the source of the brine is in the base material. Calcium chloride is sometimes used to stabilize base material. In Clayton county, Iowa, some PCC pavements, made with salt susceptible dolomite aggregates and placed over calcium chloride stabilized bases, suffered from early deterioration.
In Iowa, fine-grained dolomite aggregates do not perform well when used in portland cement concrete pavements (PCCP) that receive deicing salts. In addition to having a small grain size, these salt susceptible dolomites contain pyrite and/or detectable amounts of manganese. These fine-grained dolomites can perform adequately in county PCCP where little or no deicing salt is used. Severe, deicing salt related deterioration can be seen where brine, from the primary road system, is tracked onto the secondary county pavement. Deicing salts can reduce the PCCP service life 5 to 10 years when these salt susceptible aggregates are used in the mix.
Some fine-grained limestone (calcite) coarse-aggregates are also salt susceptible when used in PCCP. Most of the salt susceptible limestones come from quarries working bedrock 310 million years old (Pennsylvanian system). Some of the salt susceptible limestones are also quarried from ledges in the stewartville formation (Ordovician, 435 million years old). The salt susceptible limestones contain elevated trace amounts of strontium or phosphorus. The use of x-ray fluorescence (XRF) chemical testing is used to identify salt susceptible aggregates.
The extremely small amount of strontium, phosporus or manganese is most likely not directly reacting with the concrete matrix to cause early deterioration. Rather, a coating (containing one or more of these elements) on the crystal and/or crystallite surface deteriorates in the presence of brine. In PCCP, the aggregate/paste bond is broken when the salt eats through the crystal interlock that is responsible for aggregate integrity. An in-house study, using the scaning electron microscope (SEM) at Iowa State University (ISU) in Ames, Iowa, showed the crystal interlock area to be especially sensitive to salt induced corrosion. The interlocked distance may be in the range of 20 microns. Brine in the pore system, on both sides of the interlock, only has to progress 10 microns to unlock a crystal. After the crystals have lost their integrity due to the introduction of deicing salts, freeze/thaw deterioration takes over and finishes the job.
Argillaceous carbonates do not perform well in Iowa PCCP. By specification, carbonates that contain more than 6% clay can not be used for aggregate in Iowa PCC. XRF analysis can quickly and easily identify argillaceous carbonates. Thermogravimetric analysis (TGA) can also do the job.
In the future there will be other parameters (direct measurement) in carbonate aggregates that will be identified and associated with PCC failures.
When a DOT engineer wants to know why a PCCP is failing, I would hope that any petrographer would say more than its just an ACR problem.
More later-
