Predicting effect of fibers on thermal gelation of methylcellulose using Bayesian optimization

Understanding of the viscoelastic behavior of a polymer is a prerequisite for its thermomechanical processing beyond laboratory scale. Utilizing rheological characterization is a powerful tool to comprehend the complex nature and time-dependent properties of macromolecular materials. Nevertheless, it consumes time as rheometry involves iterating experiments under several conditions to visualize the non-linear behavior of materials under varying conditions. The work hereunder examines the rheology of cellulosic aqueous suspensions prepared using cellulose fibers as the dispersed phase (Refcell and Storacell) and methylcellulose (MC) as the polymeric matrix. Interfacial phenomena between MC and cellulose fibers arise in particle laden systems with supramolecular structures formed by non-covalent interactions. Therefore, this study elucidates the rheological evolution of these interactions as a function of temperature and fiber concentration. This study displays how researchers may reduce the number of rheological experiments and save time utilizing a novel method based on a Bayesian optimization with Gaussian processes.