The world’s strongest bio-based filament is currently produced from the wet spinning of cellulose nanofibrils. Such filaments could provide sustainable alternatives to currently available reinforcement fibers for high performance composites. Finding the weakest point of such filaments is of critical importance to their final application. However, what constitutes the weakest point in these filaments has yet to be determined with any certainty. Laser diffraction in conjunction with the Fraunhofer (single-slit) approximation could provide a rapid, nondestructive defectoscopy technique. Finding the thinnest point is the most accurate estimate of the breakpoint, with a mean distance-to-breakpoint of 1100 ± 200 μm, whereas the most precise determination of the weakest point is establishing which point is the least slit-like (48% of all cases). Combining width and propensity to be more or less slit-like was attempted to provide an accurate and precise metric (‘the failure factor’). With an accuracy of 1000 ± 200 μm, this is the best possible estimate using the methodology presented here. The results indicate a need for further refinement of the method. Incorporating machine learning algorithms would increase reliability by circumventing the need for approximations. Performing high-resolution tomograms of the samples to take into account the cross-sectional circularity could be implemented as an additional in-depth secondary method. With the technique already presenting advantages in terms of speed and cost compared to other methods, e.g., electron microscopy, such instruments could be integrated into a production line, providing real-time defectoscopy and quality control.
Pierce et al. (Sun,) studied this question.