Summary: | Thesis: S.M., Massachusetts Institute of Technology, Department of Mechanical Engineering, 2018. === Cataloged from PDF version of thesis. === Includes bibliographical references (pages 173-180). === In an effort to more fully leverage the relatively rich data within x-ray micro computed tomography ([mu]CT) datasets, two automated analysis tools have been developed which limit subjectivity compared to extant CT analysis of composites. The developed tools extract micron-scale morphological information in three dimensions for laminated advanced composite structures, specifically unidirectional prepreg aerospace-grade carbon fiber reinforced plastic (CFRP) laminated composites. The interlaminar thickness tool locates the interlaminar region and calculates its thickness at each voxel, and the porosity tool quantifies the void volume fraction and localizes each void to the interlaminar or intralaminar regions. Both of these tools were validated on exemplary datasets, and are shown to outperform manual methods. The interlaminar thickness measurements are accurate within 1.5 Jim and well within the standard error of the manual measurement performed via the currently accepted (destructive) method: Scanning Electron Microscopy (SEM). The tool revealed several previously unidentified features of the interlaminar region, including significant fiber misalignment at ply interfaces, and resin-rich regions that extend in the direction of ply microfibers, termed Tow-Aligned Resin Pockets (TARPs) herein. The porosity tool reveals macrovoids (volumes greater than 100 lim3) and microvoids (volumes less than 100 [mu]m 3) and localizes each void to the interlaminar or intralaminar regions. Macrovoids are studied in extant void quantification literature using optical or SEM inspection of cross-sections, but the small microvoids have been overlooked by these analyses and are found to be pervasive in the interlaminar and intralaminar regions. The tools were used to assess two aerospace unidirectional CFRP material systems: autoclave-cured AS4/8552 and out-of-autoclave (OoA) IM7/M56. Laminates of each material were manufactured with and without vertically aligned carbon nanotubes (CNTs) reinforcing the interlaminar region; this hierarchical nanoengineered architecture, termed nanostitching, is of interest in other work to improve strength and toughness of laminates. The analysis shows that the baseline AS4/8552 and IM7/M56 interlaminar thicknesses are 8.6 jrm and 14 jrm thick, respectively, and the nanostitched versions are 2.2 urm and 8.0 jim thicker than the baseline, respectively. TARPs are common to both materials, and are found to become 38 % larger with the addition of CNTs. Void content in the baseline and nanostitched materials was very low, -0.002vol% for the AS4/8552 and -0.001vol% for the IM7/M56. It was found that small microvoids (with diameters on the order of microns) are persistent in both materials, numbering -300 per mm3 with an average volume of 25 [mu]m3, and are generally evenly distributed, although some samples exhibit concentrations of small intralaminar microvoids in the certain plies. Future work will investigate possible connections of the newly-identified features (TARPs and microvoids) to mechanical properties, particularly as they may be damage precursors. Additional materials should be evaluated to determine if these newly-identified features persist in other aerospace-grade composites. In addition to these tools, this work provides a procedure for developing additional [mu]CT analysis tools, such as tools to automate damage identification and quantification. === by Nathan K. Fritz. === S.M.
|