A review of computational approaches used in the modelling, design, and manufacturing of biodegradable and biobased polymers

Abstract

The design and manufacture of new biodegradable and bioderived polymeric materials has traditionally taken place through experimentation and material characterisation. However, cutting-edge computational methods now provide a less expensive and more efficient approach to innovative biopolymer design and scale-up. In particular, the holistic framework provided by Materials 4.0 combines multiscale simulations and computational modelling with theory and next-generation informatics (big data integration and artificial intelligence) to model biopolymer structures, understand their flow and processibility, and predict their properties. These computational methods are being utilised to model and forecast the properties of a wide variety of biopolymeric materials, including the large family of biodegradable polyesters along with lignocellulosics, polysaccharides, proteinaceous materials, natural rubber, and so on. Ranging from quantum- to macroscale, computational modelling acts as a complement to traditional experimental techniques, probing molecular structure and intramolecular interactions as well as reaction mechanisms. This enables further kinetic modelling studies and molecular simulations. The research has been further expanded to include the use of machine learning approaches for material property optimisation in conjunction with expert knowledge and relevant experimental data. Aside from the modelling of structure-property relationships, computational modelling has also been used to predict the effect of biopolymer modifications and the influence of external factors such as the application of external fields or applied stress and the effects of moisture. In summary, there is a fast-developing library of computational modelling data for biopolymers, and the development of Materials 4.0 in this sector has enabled greater flexibility in design and processing options in advance of more expensive and time-consuming testing.