Residual shives have a major impact on the quality and performance of flax textile and composite preforms. Quantifying the content of shives in batches of scutched flax fibres is a time-consuming and operator-dependent process. Although the shive content can be estimated from the chemical composition, our aim here is to explore a series of effective, reliable and complementary approaches to quantifying shives. Various methods are presented: microscopic observations, analytical biochemistry (monosaccharides, lignins), indirect methods (infrared spectroscopy) and dynamic morphological analysis (QICPIC). A reference standard was first analysed and then compared with batches of flax tow fibre of unknown shive content from current industrial production. This approach shows that calibration curves can be established by applying each selected method to batches with known fibre and shive content. In addition, a database was generated to determine the shive content of industrial batches using partial least squares (PLS) regression. Finally, a detailed study of shives in industrial batches is presented, comparing the different analytical methods.
Read full abstract