Performance Limits for Evaluating Supplementary Cementing Materials Using Accelerated Mortar Bar Test

The accelerated mortar bar test (AMBT) was originally developed for the purpose of identifying alkali-silica reactive aggregates, but has been widely used to evaluate the preventive action of supplementary cementing materials (SCM). Indeed, a modified version of the AMBT for testing the effectiveness of pozzolans and slag for controlling expansion due to alkali-silica reaction (ASR) was recently developed and published as an ASTM standard test method (ASTM C 1567). In this paper, results from accelerated mortar bar tests on reactive aggregate-SCM combinations are compared with the performance of the same combination of materials in concrete structures, field-exposed concrete blocks, and laboratory expansion tests on concrete prisms (ASTM C 1293). It is concluded that the use of a 14-day expansion limit of 0.10% in the AMBT produces an outcome that agrees well with the performance of concrete in the laboratory or under field conditions. Combinations of reactive aggregates and SCM that pass this limit when tested in mortar have a very low risk of resulting in damage when used in concrete. Furthermore, the minimum level of SCM required to control expansion with a given reactive aggregate can be determined using the 14-day expansion limit and the result is in good agreement with the amount of SCM required to prevent cracking in concrete. Extending the duration of the test (for example to 28 days) is overly conservative and results in estimates of much higher levels of SCM (by 1.5 times on average) to control expansion than that actually required in concrete.

Keywords: aggregate; mortar; pozzolan; slag.

INTRODUCTION

The accelerated mortar bar test (AMBT) was originally developed in South Africa in the early 1980s as a rapid test method for identifying aggregates that could potentially cause alkali-silica reaction (ASR) when used in portland cement concrete.1 Soon after details of the test method were published,1 it was proposed that the test was also suitable for evaluating the effectiveness of supplementary cementing materials (SCM) such as fly ash, slag, silica fume, and natural pozzolans as means for preventing damaging ASR.2 Since this time there have been a number of studies demonstrating that the AMBT gives a reasonable prediction of how combinations of SCM and reactive aggregates will perform in concrete.3-6

In 2000, the Canadian Standards Association published a standard practice, CSA A23.2-28A,7 for evaluating the effectiveness of SCM in preventing expansion due to ASR, which permitted the use of the AMBT for this purpose. Combinations of reactive aggregate and SCM can be used in concrete construction provided that mortar bars produced with the same aggregate and same level of SCM expand by less than 0.10% after 14 days storage in 1 M NaOH at 80 °C (176 °F). The standard practice allows the data from the AMBT to be used until 2-year expansion data become available from the concrete prism test (CPT),7 which must be initiated at the same time. Once the CPT data are available, these results prevail.

Although the AMBT was first adopted as a means for identifying reactive aggregates by ASTM in 1994 (ASTM C 1260), a modified version for evaluating SCM was not published until 2004 (ASTM C 1567). The test method, however, has been used for this purpose by numerous agencies and the use of the test is an option in guide specifications published by both AASHTO8 and PCA.9 Both guide specifications recommend the use of a 14-day expansion limit of 0.10%* (that is, reactive aggregate-SCM combinations that expand less than 0.10% when tested in mortar bars immersed in 1M NaOH at 80 °C (176 °F) for 14 days may be used in concrete).

Thomas and Innis5 presented data from 70 different reactive aggregate-SCM combinations tested in both the AMBT and the CPT (ASTM C 1293 or CSA A23.2-14A modified by introducing SCM, as direct mass replacement of portland cement) and showed that there was a reasonable correlation between the 14-day expansion in the AMBT and 2-year expansion in the CPT. More importantly perhaps is their conclusion that combinations of reactive aggregates and SCM that expanded less than 0.10% at 14 days in the AMBT had a very low risk of exceeding the 2-year expansion limit of 0.040% in the CPT and, consequently, a low risk of expansion under field conditions.

Although a 14-day limit in the AMBT has been adopted in a number of specifications,7-9 some workers have shown a preference for extending the duration of the alkali-immersion period to 28 days and using an acceptance limit based on the expansion at 28 days.10 Such an approach has been adopted both by the Northwest Region of the Federal Aviation Administration (FAA) and the Nebraska Department of Roads (NDOR).

In this paper, the performance of different reactive aggregate-SCM combinations tested in the AMBT is compared with the performance of the similar materials in real structures and the same materials tested in large concrete blocks stored under field-exposure conditions, as well as in concrete prisms stored under laboratory conditions. The performance of concrete is used as a benchmark to determine the most appropriate performance limits of the accelerated mortar bar test.