Finite Element Analysis of Walls with Alkali Silica Reaction Subject to Simulated Seismic Cyclic Loading

Abstract of the presentation at the:
12th Canadian Conference on Earthquake Engineering (CCEE) 2019, Quebec City, QC
June 17–20, 2019

Prepared by:
Genadijs Sagals, Nebojsa Orbovic and Thambiayah Nitheanandan
Canadian Nuclear Safety Commission


The alkali–silica reaction (ASR), a type of alkali–aggregate reaction (AAR), is observed in some concrete structures in eastern Canada and the eastern United States. The only structure with ASR that is regulated by the Canadian Nuclear Safety Commission (CNSC) is Nuclear Power Plant Gentilly-2 (currently in decommissioning state). ASR-induced concrete expansion results in cracking that may degrade the mechanical properties of the concrete. The effect of ASR on the structural demand and seismic response of concrete buildings and anchors should be assessed.

The CNSC is currently developing a regulatory requirements basis for the assessment of existing concrete structures with AAR, as well as a means of avoiding this pathology in new builds. This presentation describes the research that the CNSC is conducting to predict the behaviour of an ASR wall subjected to constant axial and lateral cyclic loads. This research was conducted as part of the Organisation of Economic Co-operation and Development (OECD) / Nuclear Energy Agency (NEA) / Committee on the Safety of Nuclear Installations (CSNI) benchmark project known as Assessment of structures subject to concrete pathologies (ASCET).

The objective of this paper is to describe the use of the commercial finite-element (FE) code LS-DYNA to model concrete walls with regular concrete and reactive ASR concrete. Adequate modelling of concrete with ASR involves complex chemo-mechanical constitutive models that are outside the sets of available materials in commercial FE packages. The current work analyzes the effect of ASR in a simple phenomenological model by substituting concrete expansion due to ASR with an identical thermal expansion. Concrete strains due to ASR expansion are thus modelled as thermal strains due to a temperature increase of 1°C with a thermal expansion coefficient equal to the longitudinal concrete expansion due to ASR.

Cyclic loading with increasing amplitude was applied to both the ASR walls and the regular walls until failure was achieved. The FE predictions were compared with available test results for both the regular walls and ASR walls subjected to accelerated aging (240 days for regular walls and 260 days for ASR walls). Because there was good agreement between the FE predictions and the test results, additional FE analysis was conducted to perform a "blind" prediction of the behaviour of both the regular walls and ASR walls after 900 days of accelerated aging. Once the results of this additional test were obtained, the blind FE predictions were compared with the test results, and reasonable agreement was achieved. The FE model was revised to account for the real material data obtained in the additional test. The revised model produced good agreement for all five tests conducted: 240 and 975 days of aging for regular walls, and 260, 610 and 995 days of aging for ASR walls. Maximum shear capacity, ductility, energy dissipation, crack patterns and failure modes were considered for this comparison.

In summary, the full paper presents the detailed results obtained in the analysis, as well as the lessons learned, to provide a best-practice guideline for analyzing the effect of concrete pathologies on reinforced concrete structures.

To obtain a copy of the abstract’s document, please contact us at or call 613-995-5894 or 1-800-668-5284 (in Canada). When contacting us, please provide the title and date of the abstract.

Date modified: