This paper seeks for a better understanding of the failure mechanisms of concrete in compression and explores the possibilities of incorporating these physical mechanisms in the computational modeling of concrete behavior. In the first half, the standard compression test is revisited. It is found that failure of concrete in compression can be regarded as the sequence of a two-stage crack mechanism. First, cracks open parallel to the direction of loading, leading to the formation of small columns in the material. Second, the bending of these small columns due to both the heterogeneity of the material and the friction on the lateral edges of the small columns (friction at the lips of the vertical cracks) lead to the formation of oblique cracks (with respect to the direction of loading). The latter merge to form oblique slippage planes and result in failure. In the second half, these physical mechanisms are explored from a modeling point of view; a probabilistic numerical model is proposed that accounts for the initial heterogeneity of the material, the random creation of geometrical discontinuities (\Ii.e.\N, cracks), and the mechanisms of friction on the crack edges. Several numerical simulations show that this modeling captures well the physical mechanisms analyzed.
[1]
Wai-Fah Chen.
Plasticity in reinforced concrete
,
1982
.
[2]
Claude Boulay,et al.
Stereophotogrammetry and Localization in Concrete under Compression
,
1991
.
[3]
Claude Boulay,et al.
Influence of Boundary Conditions on Strain Softening in Concrete Compression Test
,
1993
.
[4]
F. Le Maou,et al.
Scale effect on concrete in tension
,
1994
.
[5]
Hamid Sadouki,et al.
Simulation and analysis of composite structures
,
1985
.
[6]
C. Boulay,et al.
On the importance of initial stresses in concrete and of the resulting mechanical effects
,
1987
.
[7]
P. Rossi,et al.
Numerical modelling of concrete cracking based on a stochastic approach
,
1987
.