Comparative study of a chemo–mechanical modeling for alkali silica reaction (ASR) with experimental evidences

• Published in: Construction & building materials Vol. 72; pp. 301 - 315
• Main Authors: Sanchez, L.F.M., Multon, S., Sellier, A., Cyr, M., Fournier, B., Jolin, M.
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 15.12.2014; Elsevier B.V; Elsevier
• Subjects: Alkali–silica reaction; Analysis; Chemical properties; Chemo–mechanical modeling expansion; Civil Engineering; Concrete; Deterioration; Engineering Sciences; Mechanical properties; Silica; France; Canada; Deterioration; Alkali–silica reaction; Chemo–mechanical modeling expansion; Alkali-Silica Reaction; particle size; modelling; expansion
• ISSN: 0950-0618; 1879-0526
• DOI: 10.1016/j.conbuildmat.2014.09.007

•LMDC model can evaluate both ASR kinetics and amplitude due to ASR.•LMDC model can be used for either ASR coming from reactive sands or coarse aggregates.•The impact of the scale effect on the competition between fine and coarse reactive aggregates needs to be better represented.•Consideration of microscopic features and alkalis leaching would improve the model’s reliability.•The LMDC model is able to assess both compressive and tensile strength losses due to ASR. Modeling ASR and the resulting expansions is necessary to obtain relevant predictions of the structural responses of distressed concrete elements. Thus, to be efficient and reliable, models should take into account both chemical and physical ASR aspects. The LMDC developed a chemo–mechanical micromodel which predicts concrete expansions and damages over time. However, the output results of this model were only compared to damaged mortars but never to damaged concrete cast and measured in the laboratory. This paper presents and analyzes the LMDC chemo–mechanical micromodel in regard to experimental evidences. Analyses were carried out on three types of concrete (25, 35 and 45MPa) incorporating two different highly-reactive aggregates (New Mexico gravel and Texas sand). Both the assumptions and the input/output parameters used/obtained in/from the model are discussed. The results showed that the LMDC modeling is reliable and effective in the ASR expansion/damage predictions.