Discrete volumetric digital image correlation for the investigation of granular type media at microscale: accuracy assessment

The recent development of efficient 3D imaging tools such as X-Rays computed microtomography combined with the extension to volumetric images of Digital Image Correlation (DIC) techniques provide new insights on the analysis of materials and structures. Among many other possible fields of application, geomaterials are good candidates for such investigations, owing to their relative transparency to X-rays and the presence in many samples of a natural contrast suitable for deformation mapping. However, these materials often deform discontinuously at microscale, for instance in the form of the development of a networks of microcracks. Discontinuity is even the dominant rule in granular-type materials such as sand in which the contribution to overall deformation of the microcontinuous phenomena-elastic strains inside grains-are negligible. To investigate deformation at the scale of these discontinuous mechanisms, specific DIC algorithms are required, which override the assumption of continuity of the transformation at the scale of the correlation windows. The recent so-called Discrete-DIC procedure (Hall et al, 2010) is a possible answer. We recall here its general principles and focus on its potential accuracy, from both theoretical and practical points of view. We show that the position and the rotation of individual grains with an average diameter of 500 mu m can be determined from images recorded with a laboratory microCT scanner, with a 15 mu m voxel size, with an accuracy of the order of 1 mu m and 0,1 degree, respectively.