Numerical strategies for prediction of drying cracks in heterogeneous materials: Comparison upon experimental results

• Published in: Engineering structures Vol. 33; no. 3; pp. 920 - 931
• Main Authors: Lagier, F., Jourdain, X., De Sa, C., Benboudjema, F., Colliat, J.B.
• Format: Journal Article
• Language: English
• Published: Kidlington Elsevier Ltd 01.03.2011; Elsevier
• Subjects: Aggregates; Applied sciences; Building failures (cracks, physical changes, etc.); Buildings. Public works; Cement concrete constituents; Cement paste; Cements; Civil Engineering; Computation methods. Tables. Charts; Cracking; Drying; Durability. Pathology. Repairing. Maintenance; Engineering Sciences; Exact sciences and technology; Finite Elements simulations; Fracture mechanics; Materials; Mathematical models; Mesoscopic scale; Pastes; Properties and test methods; Properties of anhydrous and hydrated cement, test methods; Shrinkage; Strain incompatibilities; Structural analysis. Stresses; Mesoscopic scale; Finite Elements simulations; Strain incompatibilities; Cement paste; Cracking; Shrinkage; Drying; Deformation; Forecast model; Finite element method; Experimental result; Concrete construction; Numerical simulation; Comparative study; Aggregate(materials)
• ISSN: 0141-0296; 1873-7323
• DOI: 10.1016/j.engstruct.2010.12.013

Both experimental and numerical studies are presented dealing with the separation, identification and quantification of drying shrinkage incompatibilities effects between cement paste and aggregate on mechanical behaviour of cementitious materials. The experimental protocol allows us to separate these effects due to the material heterogeneity from the structural ones (drying shrinkage gradient). In order to limit the parametric study to size and volume fraction effects, a “model” (controlled) material, composed of cement paste and cylindrical sandstone aggregates, is considered. Cracking observation is undertaken using digital image correlation. These experimental investigations have been completed by a numerical study to evaluate the ability of two different models (using either smeared or discrete cracks description) to describe the experimental behaviour observed at the mesoscopic scale. [Display omitted]