Detecting damage initiation in short fiber composites via in-situ X-ray tomography and digital volume correlation

Abstract

In-situ X-ray tomography is an effective means for the acquisition of time-resolved 3D images of fiber reinforced composites during mechanical loading and deformation. However, the manual comparison of corresponding in-situ X-ray images at successive loading instances to detect damage and its propagation is a time-consuming process. To improve the detection of damage using in-situ X-ray tomography, digital volume correlation (DVC), a full field 3D-strain technique that determines relative deformation between consecutive image volumes, was used. Specifically, regions of high strain were isolated from the heterogeneous strain computed from DVC within the microstructure and compared to manually detected damage. In this study, 59 out of 63 manually detected damage events, which included fiber breakage, micro-void nucleation, and fiber debonding, were within the search spaces of high strains beyond the 60th percentile threshold computed by DVC. Results of this study demonstrate the use of DVC as a tool to confine volumes of potential damage sites within the in-situ X-ray tomograms of composites to improve the efficiency of incipient damage detection.