Skip to contentUnited States Department of Transportation - Federal Highway AdministrationSearch FHWAFeedback



Alkalis (Na, K, and Ca) are imperative to ASR. The Na+ and K+ ions combine with the "freed" silica to form ASR gel; and the Ca++ ions play a key role in the expansion of the gel (How exactly Ca++ effects gel expansion is unclear, but there are many investigations that suggest the absence of Ca++ ions would cause the gel to be non-expansive).  Without a sufficient supply of Na+ and K+ ions, the pore solution pH drops, the solubility of silica is reduced, and ASR gel formation hindered.  Without Ca++ ions, the gel would not expand and deleterious deterioration would not occur. 
Sources of alkalis in concrete include:

  • Portland cement
  • Other cementing materials
    • Fly ash
    • Slag
    • Silica fume
  • Chemical admixtures
  • Wash water (if used)
  • Aggregates
  • External sources
  • Seawater
  • Deicing chemicals
  • Portland cement

Alkalis in cement

The chemical make up of regular Portland cement includes sodium and potassium in the form of Na2O and K2O. Therefore, an appreciable amount of the necessary alkalis for ASR to occur are supplied by the cement.  Calcium, in the form of CaO, is also present.  However, it is not considered when calculating the alkali content of cement.

Figure A2a.F1. Illustration. Alkalis in Cement are Represented in an Oxide Analysis Table. This illustration shows an oxide analysis table on the right side. Sodium and potassium ions are circled in red with an arrow that points to writing on the upper left of the illustration. Below the writing appear 2 chemical formulas describing the generation of hydroxyl ions. The hydroxyl ion generation is circled in blue and an arrow extends from the circle to an explanation of how pH is affected.

Figure A2a.F1. Alkalis in Cement are Represented in an Oxide Analysis Table.

In order to calculate the total alkali contribution of cement, the amount of K2O is converted to "equivalent sodium" by way of multiplying the reported amount by 0.658 (the ratio of molecular weight of K2O to Na2O) and combining it with the Na2O amount.  The alkali equivalent, or equivalent soda, is calculated using the following equation:

Alkali Equivalent or "Equivalent Soda":

Na2Oe = Na2O + .0658 K2O

Figure A2a.F2. Illustration. Example of Calculating Total Alkalis in Cement. This illustration shows an oxide table on the left. On the right is the mathematical solution for the total alkali equivalent.
Figure A2a.F2.  Example of Calculating Total Alkalis in Cement.

In 1995, Gebhardt showed that for one type of cement the chemistry could be very different depending on the manufacturer.  He showed that for 69 North American Type I cements tested, just over half had alkali contents less than 0.60%.  The other half had greater alkali contents.

Figure A2a.F3. Illustration. Alkali Contents in 69 Type I Cements. This illustration shows the results of a research study done on 69 typical Type I cements found in North America. The results are depicted in a bar graph on the right side of the illustration. A description, the author, and year of publication are listed on the right side.
Figure A2a.F3. Alkali Contents in 69 Type I Cements.

In 1940 and 1952, Stanton showed that alkali content influenced the amount of expansion measured in his mortar bar testing method.  He found that the more alkalis in a cement, the greater the expansion.  The graph in Figure 5.9 shows Stanton's results.  This work defined the concept of "low-alkali cement".  It is still generally believed today (wrongly so) that the use of such cement only is a sufficient measure to avoid damaging ASR.  Effective measures in avoiding ASR are discussed in the section titled Controlling ASR.

Figure A2a.F4. Illustration. Results of Stanton’s Mortar Bar Tests. This illustration shows a graphical depiction of results on the right side. On the right side is an explanation of the likelihood for expansion using cements with alkali contents less than 0.6% Na2Oe.
Figure A2a.F4. Results of Stanton’s Mortar Bar Tests. 

Alkalis in concrete

Although alkalis supplied by cement play an important role, it is now fairly well established that it is the alkali content in the concrete and not just the alkali content of the cement that influences the risk of damaging reaction for a particular aggregate. The concrete alkali content is calculated as follows:

Figure A2a.F5. Illustration. Calculation of Alkalis in Concrete. The illustration shows the equation for calculating concrete alkali content in both metric and SI units.
Figure A2a.F5.  Calculation of Alkalis in Concrete.

Figure A2a.F6. Illustration. Alkalis example. This illustration shows how the equation for calculating alkalis in concrete is applied.
Figure A2a.F6. Alkalis example.

Work done in Canada (Figure 5.12) shows that after two years, concrete containing alkalis at less than 3.0 kg/m3 (5lbs/yd3) had the potential of staying below the critical threshold limit set by Canadian Standards Association (CSA).  Expansions of concrete with greater alkali content were well above the threshold limit and could potentially result in deleterious expansion in the field.  The threshold alkali content - above which deleterious expansion occurs - varies with the type of reactive aggregate.

Figure A2a.F7. Illustration. Effect of Concrete’s Alkali Content on Expansion. This illustration shows a graphical representation of results on the right side. Values below CSA values are circled in blue and explained on the left side of the illustration.
Figure A2a.F7.  Effect of Concrete’s Alkali Content on Expansion. 


Updated: 02/20/2015

United States Department of Transportation - Federal Highway Administration