A kinematic measurement for ductile and brittle failure of materials using digital image correlation

Abstract

This paper addresses some material level test which is done on quasi-brittle and ductile materials in the laboratory. The displacement control experimental program is composed of mortar cylinders under uniaxial compression shows quasi-brittle behavior and seemingly round-section aluminum specimens under uniaxial tension represents ductile behavior. Digital Image Correlation gives full field measurement of deformation in both aluminum and mortar specimens. Likewise, calculating the relative displacement of two points located on top and bottom of virtual LVDT, which is virtually placed on the surface of the specimen, gives us the classical measure of strain. However, the deformation distribution is not uniform all over the domain of specimens mainly due to imperfect nature of experiments and measurement devices. Displacement jumps in the fracture zone of mortar specimens and strain localization in the necking area for the aluminum specimen, which are reflecting different deformation values and deformation gradients, is compared to the other regions. Since the results are inherently scattered, it is usually non-trivial to smear out the stress of material as a function of a single strain value. To overcome this uncertainty, statistical analysis could bring a meaningful way to closely look at scattered results. A large number of virtual LVDTs are placed on the surface of specimens in order to collect statistical parameters of deformation and strain. Values of mean strain, standard deviation and coeffcient of variations for each material are calculated and correlated with the failure type of the corresponding material (either brittle or ductile). The main limiters for standard deviation and coeffcient of variations for brittle and ductile failure, in pre-peak and post-peak behavior are established and presented in this paper. These limiters help us determine whether failure is brittle or ductile without determining of stress level in the material.