Subject matter

The overall goal of the proposed TC is to develop a framework for risk assessment of mixture designs for concrete prone to alkali-silica reaction (ASR). This includes validating current test methods for aggregate reactivity, efficacy of SCMs and alkali balance in the concrete system.  The data from these validation efforts (WP 1 and WP 2) will inform a framework (WP 3) that is a starting point for the development of decision-making models. This framework would allow the user to enter known data about a particular aggregate-cementitious materials composition, concrete alkali content from various (external/internal) sources, the expected exposure environment, service-life needs and structural classification and to determine a pathway for mixture designs with reduced risk for deleterious ASR. 

Assessing aggregate reactivity alone is no longer particularly meaningful in the absence of considering the context of the entire concrete system.  The expansion behaviour by ASR can be different depending on aggregate character (rock type and its combination, size, alkali-absorbing/releasing properties, mix proportion), alkali inventory and age of the concrete. The framework should encompass the entire concrete system (WP1 – aggregate reactivity; WP2 – alkali inventory).

In addition, the framework should account for the environmental conditions to which the concrete is exposed. Hereto, the access to field stations with a large variety of climates is crucial (WP1). As the world is currently facing climate change, exposure conditions can change rapidly. In the long-term the framework should enable estimations of ASR expansion at various environmental conditions.

Preventive measures currently adopted are established based on accelerated laboratory conditions that do not consider any temperature/moisture cycling and bench-marked against numerous field experiences. Further validation of these preventive measures is needed, including a time scale and variation of effectiveness in different environments [1]. This will be included in the framework (WP3).  

The group assembled for this TC has a very strong tie to their respective national standards organizations (e.g. AASHTO, ASTM, ACI, CSA, JSA, CEN, FIB and others).  Many of these members are active participants and leaders in the organizations.  This ensures strong cross-collaboration between the groups and where possible sharing of knowledge and improvement to test methods and standards. 

1. Rajabipour, F., Giannini, E., Dunant, C., Ideker, J.H., and Thomas, M.D.A., Alkali–silica reaction: Current understanding of the reaction mechanisms and the knowledge gaps. Cement and Concrete Research, 2015. 76: p. 130-146.

Terms of reference

The proposed work will fill the normal 5-year lifespan of a TC (see Table 1). Given the current COVID-19 situation it is proposed that a first meeting be held virtually in the Spring of 2021 and the second meeting will be held in Mexico commensurate with the RILEM week in the Fall of 2021.  The TC comprises 3 work packages (WP). It is proposed that WP 1 will take place in years 1-3, WP 2 will occur in years 1-4 and WP 3 will take place in years 3-5 of the proposed TC. A proposed list of 46 members with broad global representation (all interested, helped in proposal development and committed to the TC) is provided in Section 9.  Several Postdoctoral Scholars are included in this list.  We have recruited 5 (five) new members to this TC who are also new to RILEM.  We will be open and will invite other current RILEM members who have ongoing projects and/or expertise in ASR as well.  In this TC there will be literature-based research, new information coming from the various ongoing and/or new projects outlined in Section 10 as well as development of a framework to be the basis for a decision-making model.  No round-robin testing is anticipated as part of this TC. The industrial relevance for this TC will be in the form of validation/verification of existing RILEM test methods for ASR determination (reactivity) and prevention and particularly in the framework that will be developed.  This framework will establish a new protocol for selecting mixture designs that will be resistant to ASR that includes factors such as exposure, structural classification and lifespan of the concrete structures.  This is not currently done and is of great interest to the industry.

Detailed working programme

The goal of the proposed TC is to develop a framework for risk assessment of mixture design for concrete containing alkali silica-reactive aggregates.

Central in such a framework is the determination of the uncertainty of the selected accelerated performance-based test methods by comparing to field performance (WP1), as well as systemizing the impact of the different alkali sources on both reactivity and prevention (WP2).    

The  framework (WP3) would provide the user with a risk-assessment process relying on known information about a particular aggregate-cementitious materials composition, its expected exposure environment, service-life needs and structural classification (risk of ASR resulting in deleterious expansion) to determine a mixture design and risk levels associated with the potential for deleterious alkali-silica reaction.

Figure 1 gives a schematic overview of the work packages and how they are connected. The timeline for the TC is presented in Table 1. In the following sections the activities in the different work package are explained in further detail.

Figure 1 : Schematic overview of the work program of the TC

WP 1 Validation of Accelerated Laboratory Methods against Field Performance and Outputs Interpretation

Now that a mature suite of RILEM test methods exist (following the work of previous TCs) the user needs guidance on how to choose which RILEM method (s) is (are) appropriate for the project under consideration and how this interfaces with local/national guidance/standards/test methods.  Not only does a user need to know which methods to choose but what the implications of a result are from that test method and how test results from different methods should be interpreted.

An inventory of relevant existing field structures and exposure sites was compiled during TC 258-AAA. As the exposure sites are maturing, we are able to gain valuable information to further benchmark and refine our interpretation of accelerated test methods. Benchmarking will for example provide crucial information regarding the reliability of laboratory methods to predict preventive measures that have successfully and unsuccessfully prevented ASR in the field. This will include traditionally used SCMs such as fly ash, slag, silica fume as well as lightweight aggregates, natural pozzolans, ground glass and calcined clays. 

We will investigate and compare the reliability of accelerated test methods for the determination of the reactivity level of an aggregate, and efficacy of preventing effect of SCMs. It is known that the current available and recommended methods do not necessarily provide the same assessment of the alkali-reactivity potential of concrete aggregates or preventive effect of SCMs [2, 3]. When comparing the different test methods, we will focus on the uncertainty associated with the performance test methods as well as the time required for testing.

To further support the work in WP1 valuable information will provide from a series of projects as well as access to data from exposure sites. The involved Project Investigators are interested and committed to being involved in the proposed TC.

Table 2 : Projects that will provide input to WP1

Deliverable:  A report on the comparison of the reliability of a selection of performance test methods with regard to aggregate reactivity and preventive impact of SCMs will be produced. The report will contain formal recommendations to AAR-0 and revisions to and/or new test methods will be made. The report will be included as a chapter in the overall STAR produced from this TC.

WP 2 Alkali Inventory in Concrete and Impact on Reactivity and Prevention

Essential work was started in TC258-AAA to develop a test method (RILEM AAR-8) to assess alkali release by aggregates.  The test method was developed and a round robin study was completed.  Now, it is important to determine the implication of the results from this test method in concert with the total alkali inventory from the concrete (alkalis from cement, SCMs, aggregates, chemical admixtures, potential for external sources) and the important role of alkali threshold needed to initiate ASR which varies based on aggregate mineralogy/reactivity.  To accomplish WP 2 the following tasks are envisioned:

Task 1:  Assessing the implication and interpretation of results from the test method RILEM AAR-8 (2020) Determination of Potential Releasable Alkalis by Aggregates in Concrete. The remaining question from RILEM AAR-8 is “how does the user interpret the results of the test?” This links in part to WP 1 in terms of the reliability and selection of the proper accelerated test method (s). We will look into the impact of releasable alkalis from both reactive and non-reactive aggregates [4].

Task 2:  Determining the threshold alkali content for specific aggregate types to initiate reaction. We will be looking into a standard test method to assess this e.g. RILEM AAR-10 and/or AASHTO T380 Miniature Concrete Prism Test, at different alkali contents (pH in the range 13 to 13.7). Mortar bar tests should not be used. We will contribute to build a database of alkali thresholds for different aggregate mineralogy.

Task 3:  Quantifying and determining the impact of SCMs on the alkali inventory with a focus on fly ash, slag, natural pozzolans, calcined clays and ground glass, by focusing on pore solution data from existing and new studies on concrete, mortar and paste [5, 6]. Experimental data will be combined with thermodynamic modelling (GEMS).  

Task 4:  Assessing the significance of external sources of alkalis (potential modification to RILEM AAR-12) such as de-icing salts used on road pavements or de-icers used on airport pavements.

To further support this work several projects can provide valuable information to WP2 and those Project Investigators are interested and committed to being involved in the proposed TC.

Table 3 : Projects that will provide input to WP2

Deliverable:  A report summarizing findings from literature and ongoing projects related to alkali threshold, impact of SCMs on the alkali inventory and external alkali will be produced. The report will in addition contain formal recommendations for the interpretation of the results from RILEM AAR-8. The report will be included as a chapter in the overall STAR produced from this TC.

WP 3 Framework for Risk Assessment of Mixture Designs Incorporating Alkali-silica Reactive Aggregates

In this Work Package the information developed in WP 1 and WP 2 will directly feed the development of a framework for risk assessment of mixture design for concrete containing alkali-silica reactive aggregates. The framework will link the following parameters:

• The uncertainty from different AAR test methods (RILEM, ASTM, CSA, JSA, etc.) and impact of minor differences between the test methods (WP 1)
• The level of aggregate reactivity (WP 1)
• The chemical composition of the binder and the alkali inventory of the concrete (WP2)
• Role of environment in the expected field conditions of the structure to be built, with a particular focus on temperature and moisture (rainfall and/or RH) (take from WP1, and TC 258-AAA STAR)
• Structural classification (adapt from ASTM C1778, AASHTO R80) and service life

The development of the framework can encompass:

• Treatment of large data sets, data mining, and deep learning which may attract new RILEM members.
• Material/Structural modelling approach for more explicit risk assessment [7]

During the development of the framework we will coordinate with the TC led by Menéndez Méndez for data from that TC which may be useful in refining the framework.

In the long term, the framework could provide a base for probabilistic risk assessment of the influence of climate change on ASR expansion, by combining numerical modelling and climate data (optimistic/pessimistic scenario available from IPCC).

Projects that can support WP3 and in which proposed TC members are involved are reported in the table below.

Table 4 : Projects that will provide input to WP3

Deliverable:  A report and set of flow charts presenting the framework for performing risk assessment of mixture design comprising the key parameters from WP1 and WP2. The report will be included as a chapter in the overall STAR produced from this TC.

Final TC Deliverable:  A STAR that incorporates Work Packages 1-3 culminating in the development of the framework for risk assessment of mixture designs.  Further, formal recommendations on revisions RILEM AAR-0 and AAR-8, and potentially other RILEM AAR test methods will be provided.

2. Thomas, M.D.A., Fournier, B., and Folliard, K.J., Alkali-aggregate reactivity (AAR) facts book. 2013, Federal Highway Admininstration (FHWA), US Department of Transportation. p. 212.
3. Golmakani, F. and Hooton, R.D., Impact of pore solution concentration on the accelerated mortar bar alkali-silica reactivity test. Cement and Concrete Research, 2019. 121: p. 72-80.
4. Einarsdottir, S.E and Hooton, R.D., Factors Affecting the Alkali Release from Concrete Aggregates. in Proceedings, 16th International Conference on Alkali-Aggregate Reaction, . 2021. LIsbon, Portugal.
5. Vollpracht, A., Lothenbach, B., Snellings, R., and Haufe, J., The pore solution of blended cements: a review. Materials and Structures, 2016. 49(8): p. 3341-3367.
6. Fily-Paré, I., To Be Published: Behaviour of concrete incorporating Ground Glass (GG) and alkali reactive aggregates, in Département de géologie et de génie géologique. 2020, Université Laval: Québec, Canada.
7. Liu, Q.-f., Shen, X., Jiang, W.-Q., Xia, J., Hou, D., Hu, Z., and Li-Xuan, M. Coupled Deterioration Mechanisms under Various Degradation Processes. 2020 [cited 2020 08/09/2020]; Available from: https://www.researchgate.net/project/Coupled-Deterioration-Mechanisms-under-Various-Degradation-Processes.