Assessment of accuracy of fibre orientation measurement using X-ray computed tomography


D. Salaberger, P. Hine, M. Jerabek, J. Kastner - Assessment of accuracy of fibre orientation measurement using X-ray computed tomography - Proceedings of the 20th International Conference on Composite Materials , Copenhagen, Dänemark, 2015, pp. 8


Non- destructive testing using X-ray computed tomography (CT) has proven to be a powerful tool for qualitative and quantitative characterisation of filled polymers. The question that has not been addressed yet is the accuracy of the published quantifications. This work deals with the 3D characterisation of glass fibres in a polypropylene matrix. The focus lies on the determination of fibre orientation by means of tensor representation. High resolution CT was applied using a laboratory system together with an in- house developed data analysis tool. The paper shows results about accuracy of the developed method by comparing the results with ground truth data. In a second part the method is compared to the destructive standard method of sectioning and light optical image analysis. High accuracy of more than 93% was observed at 2 µm voxel edge length. The influence of resolution is higher on length distribution but very small on orientation within the chosen range of voxel edge lengths. For the comparison of 2D and 3D method the orientation tensor elements are shown as values over sheet thickness at 21 positions. The samples show a distinct core-shell layered structure. By visualizing the tensors as ellipsoids, it is possible to get an impression about degree and direction of orientation. Both methods deliver almost the same values at every position. Values for mean tensors of the complete volumes under investigation show deviations below 0.007. The results indicate that CT delivers accurate and valid results of fibre orientation. The advantages compared to the standard methods is the non- destructive manner, the volumetric analysis, the complete tensor representation and the fact, that additional information like fibre length distribution can be determined from the on single analysis.