Skip to contentU.S. Department of Transportation/Federal Highway Administration
Pavement | Bridge
FHWA > Pavement : Bridge > ASR > HIF-09-001 > 3.0 Testing To Determine Aggregate Reactivity
<< Previous Contents Next >>

Report on Determining the Reactivity of Concrete Aggregates and Selecting Appropriate Measures for Preventing Deleterious Expansion in New Concrete Construction

3.0 Testing To Determine Aggregate Reactivity

Figure 2 shows a flow chart of the sequence of testing for this report when evaluating the reactivity of an aggregate source. Selecting an appropriate preventive measure will be dealt with in the next section. Interpreting the results from these aggregate tests is discussed at the end of this section (see Section 3.6 below).

3.1. Use of Field Performance History

The long–term field performance history of an aggregate can be established by conducting a survey of existing structures that were constructed using the same aggregate source. As many structures as practical should be included in the survey and these structures should, where possible, represent different types of construction (pavements, sidewalks, curb and gutter, elements of bridges, barrier walls and even non–transportation structures). The following information should be collected for each structure:

  • Age – structures should be at least 10 years old and preferably more than 15 years old as dam age due to AAR can take more than ten years to develop.
  • Cement content and alkali content of the cement used during construction.
  • Use of pozzolans or slag during construction.
  • Exposure condition – availability of moisture, use of deicing chemicals.
  • Presence of symptoms of distress due to AAR or other causes.

Cores should be taken from a representative number of these structures and a petrographic examination be conducted in accordance with ASTM C 856 to establish the following:

  • The aggregate used in the structures surveyed is close in mineralogical composition to that of the aggregate currently being produced.
  • There is no evidence of damage due to AAR.
  • The presence and quantity of fly ash or slag.

If the results of the field survey indicate that the aggregate is non–deleteriously reactive, the aggregate may be used in new construction provided that the new concrete is not produced with a higher cement alkali loading, lower amount of pozzolan or slag, or more aggressive exposure condition than the structures included in the survey.

There is a certain level of risk associated with accepting aggregates solely on the basis of field performance due to difficulties in establishing unequivocally that the materials and proportions used more than 10 to 15 years ago are similar to those to be used in new construction.

Flowchart. Sequence of Laboratory Tests for Evaluating Aggregate Reactivity. This flowchart describes the process of evaluating the reactivity of an aggregate source. The approach is described in Section 3.6.
Figure 2. Sequence of Laboratory Tests for Evaluating Aggregate Reactivity

The type of reaction only needs to be determined after the concrete prism test if the aggregate being tested is a quarried carbonate that has been identified as being potentially alkali-carbonate reactive by chemical composition in accordance with test method CSA A23.2-26A
Note: The heavier dotted lines represent the preferred approach whereas the faint dotted lines represent a higher risk approach.

If field performance indicates that an aggregate source is deleteriously reactive, laboratory expansion testing is required to determine the level of aggregate reactivity and to evaluate prevention measures.

Petrographic Assessment

Petrographic examination of aggregates should be conducted in accordance with ASTM C 295. Petrography can reveal useful information about the composition of an aggregate, including the identification and approximate quantification of reactive minerals. Petrography may be used to classify an aggregate as potentially reactive, but expansion testing is required to determine the extent of the reactivity and appropriate levels of prevention. Aggregates may be accepted as non–reactive solely on the basis of petrography but there is certain level of risk associated with such a decision as some reactive phases may not be detected by optical microscopy (e.g. finely dispersed opaline silica found in some siliceous limestones). Where a trained petrographer is examining aggregates from well–known and tested sources, it is acceptable to use petrography to classify the aggregate reactivity on a routine basis. For example, in deposits where chert is known to be the only reactive component and where testing has shown that the chert content needs to exceed 5% to cause deleterious reaction, it may be justified to permit acceptance of an aggregate with less than say 3% chert on the basis of petrography.

In addition to looking for alkali–silica reactive minerals such as opal, chalcedony, cristobalite, tridymite, strained and micro–crystalline quartz, and volcanic glass, petrographers should also be vigilant as to the presence of mineralogical and textural features characteristic of alkali–carbonate rocks. Deleterious alkali–carbonate reactive rocks are often characterized by a microscopic texture consisting of dolomitic rhombs floating in a fine–grained matrix of calcite, quartz and clay. However, there have been reports of deleterious ACR with rocks that do not exhibit this "classic" texture (Ozol, 2006).

Determination of Potential Alkali–Carbonate Reactive Rocks by Chemical Composition, CSA A23.2–26A

If the aggregate being assessed is a quarried carbonate rock, the potential for alkali–carbonate reaction may be assessed on the basis of its chemical composition (Rogers, 1986). This is the basis for the test method CSA A23.2–26A.2 This test involves the determination of the lime (CaO), magnesia (MgO) and alumina (Al2O3) content of the rock, and determining where the composition of the rock falls on a plot of CaO/MgO ratio versus the Al2O3 content, as shown in Figure 3. If the composition falls in the range of "aggregates considered to be potentially expansive," the aggregate is potentially alkali–carbonate reactive. Such aggregates must be tested using the ASTM C 1293 (concrete prism test) as ASTM C 1260 (accelerated mortar bar test) is not suitable for detecting alkali–carbonate reaction.

Graph. Using Chemical Composition as a Basis for Determining Potential Alkali-Carbonate Reactivity of Quarried Carbonates (from CSA A23.2-26A). This graph shows rock cylinder expansion test and concrete prism expansion test results for quarried carbonates. The x-axis of the chart displays the alumina content in percent, and the y-axis displays the ratio of lime to magnesia. The graph shows a scattered plot of data points from the rock cylinder expansion tests and the concrete prism expansion tests. Two lines are plotted in this graph, and any points that fall within these two lines represent aggregates that are considered potentially expansive; data points that fall outside these two lines are considered non-expansive.
Figure 3. Using Chemical Composition as a Basis for Determining Potential Alkali–Carbonate Reactivity of Quarried Carbonates (from CSA A23.2–26A)

3.4. Accelerated Mortar Bar Test, ASTM C 1260

If the aggregate is not a quarried carbonate or it is a quarried carbonate with a composition that falls outside of the region of "aggregates considered to be potentially reactive" in Figure 3 when tested in accordance with CSA A23.2–26A, the next step is to test the aggregate in accordance with ASTM C 1260.

Coarse aggregates for this test have to the crushed to sand size (< 5 mm) and then washed and graded to meet the grading requirements of the test. Sands have to be washed and graded to meet the same grading requirements. The test is intended to evaluate coarse and fine aggregates separately, and should not be used to evaluate job combinations of coarse and fine aggregates.

In this test, mortar bars are produced with the aggregate being evaluated and at the age of two days the bars are immersed into a solution of 1 M NaOH maintained at a temperature of 176°F (80°C) and the length change of the bars is monitored. If the mortars do not expand by more than 0.10% after 14 days immersion in this solution, the aggregate is considered non–deleteriously reactive. If the mortar bar expands by more that 0.10% at 14 days, the aggregate is considered to be potentially reactivity and its reactive should be confirmed in ASTM C 1293 (concrete prism test).

3.5. Concrete Prism Test, ASTM C 1293

The concrete prism test is suitable for evaluating all aggregate types and is considered to be the most reliable laboratory test for predicting field performance of aggregates. If the aggregate being tested is a coarse aggregate it is blended with a non–reactive fine aggregate and vice–versa, and the coarse–fine aggregate combination is used to produce concrete prisms with a specified high alkali loading. The test is not intended for use with specific job combinations of coarse and fine aggregate, however, it is generally considered acceptable to do this, but the results are not applicable if either the coarse or fine aggregate is changed during the job.

The prisms are stored over water in sealed containers at 100°F (38°C) and the length change is monitored periodically. If the prisms do not expand by more than 0.04% after 1 year, the aggregate is considered non–deleteriously reactive and may be used in concrete with no further testing (for AAR). If the prism expands by more that 0.04% at 1 year, the aggregate is considered to be potentially reactive and preventive measures are required if the aggregate is to be used in concrete construction.

If the aggregate tested was a quarried carbonate rock with a chemical composition that fell within the region of "aggregates considered to be potentially reactive," the concrete prisms must be examined to determine whether alkali–carbonate reaction contributed to the expansion.3 If damaging ACR is detected, either in isolation or in combination with ASR, the rock should not be used in concrete without selective quarrying or aggregate beneficiation to remove the reactive components.

3.6. Interpretation of Results from Laboratory Tests

Figure 2 shows dotted lines from the boxes marked "Field History", "Petrographic Examination", and "Accelerated Mortar Bar Test" to the box marked "Non-Reactive" as there is an element of risk associated with accepting aggregates solely on the basis of these tests. With field history it is usually difficult to firmly establish that the materials and conditions to be used in a new project are the same as those used in a structure that is more than 10 to 15 years old. With some aggregates it may not be possible to identify reactive constituents by petrographic examination.

The accelerated mortar bar test (AMBT) is generally recognized as a relatively severe test and it is well established that it identifies as deleteriously reactive many aggregate sources that have a history of satisfactory field performance and that perform well in the concrete prism test (CPT);4 that is the CPT identifies the same aggregate as non-deleteriously reactive. For this reason, results from the accelerated mortar bar test should not be used to reject an aggregate. If an aggregate fails the AMBT (expansion > 0.10% at 14 days) its reactivity should be confirmed by testing using the CPT. If the results of the AMBT and CPT are in disagreement, the results of the CPT shall prevail.

Until recently it was assumed that aggregates that passed the AMBT (expansion ≤ 0.10% at 14 days) were most likely to pass the CPT (expansion ≤ 0.04% at 1 year), and such aggregates could be accepted for use in concrete without the need for confirmatory testing using the CPT. However, there appears to be an increasing number of coarse aggregates that pass the AMBT and fail the CPT (Folliard et al. 2006) and this is somewhat disconcerting for specifications that permit the use of aggregates passing the AMBT with no further testing (that is no requirements for CPT).5 Consequently, there is a risk associated with accepting an aggregate solely on the basis of the results from the AMBT.

In Figure 2, the AMBT is shown with a broken line as there is the possibility of incorrectly identifying a deleteriously–reactive aggregate as being non–deleteriously–reactive using this test method. The most reliable approach for determining aggregate reactivity is to use the CPT as the expansion test for all cases (that is to exclude the AMBT from the evaluation process). However, it is recognized that the long duration of the CPT makes it impractical for use in many circumstances and there is a need for a more rapid test. Despite its limitations, the AMBT is probably the most viable accelerated test currently in use.

3.7. Risk Evaluation

The risk of AAR–damage occurring due to a failure to detect deleteriously reactive aggregate can be reduced by implementing routine testing using petrography and/or laboratory expansion tests. Increasing the complexity and frequency of testing will result in lower risks but higher costs. For example, frequent petrographic and concrete prism testing of all aggregate sources may reduce the risk of failing to identify deleteriously reactive aggregates to a negligible level, but the costs associated with this level of testing may not be justified in a region where there are few cases of AAR and where most aggregate sources have a good field performance history. It is incumbent on the owner to define what level of risk is acceptable and thus determine the type and frequency of testing.

3.8. Other Test Methods

There are many other test methods for evaluating the reactivity of aggregates and a full discussion of these tests is beyond the scope of this report. One of the more promising tests is an accelerated version of the CPT, which is conducted at 140°F (60°C) to accelerate the reaction. A detailed discussion of these tests can be found elsewhere (Thomas et al. 2006).


2 CSA is the Canadian Standards Association. CSA Standards such as A23.2 Method of Test and Standard Practices for Concrete can be found at www.csa.ca

3 The determination of the extent to which the alkali–carbonate reaction contributed to the expansion of the concrete should be conducted by an expert with experience of ACR. Methods used might include a petrographic examination of the concrete (ASTM C 856), accelerated microbar testing of the aggregate (Lu et al., 2004), and/or rock cylinder expansion tests (ASTM C 586) conducted on samples of rock from the quarry. The ASTM C 1105 version of the concrete prism test for alkali–carbonate rock reaction may also be used but the alkali content of the concrete should be kept sufficiently low to ensure that expansion due to alkali–silica reaction is eliminated during the test. Keeping the alkali content below 1.8 kg/m3 (3.0 lb/yd3) Na2Oe should be sufficient for this purpose.

4 The term "false positive" is used to describe the case where a test method incorrectly identifies an aggregate as deleteriously reactive. Similarly, "false negative" describes the case where a test wrongly identifies an aggregate as being non–deleteriously reactive.

5 It has been proposed to extend the duration of this test and to use an expansion limit of 0.10%, or even, 0.08% after 28 days immersion in 1 M NaOH at 176 °F (80 °C). This more onerous requirement should be adopted only when it can be demonstrated that extension of the test period is required to capture aggregates that have been identified as being deleteriously reactive either by concrete prism testing or field performance. The extended test duration should not be applied across the board to all aggregates as this will result in an unacceptable number of cases where the accelerated mortar bar test results in false positives (that is the test wrongly identifies aggregates as deleteriously reactive). It should be noted that extending the test duration does not capture all of the aggregates that have been found to pass the AMBT but fail the CPT.

<< Previous Contents Next >>
PDF files can be viewed with the Acrobat® Reader®
Updated: 06/13/2012