Boosting structural food science using X-ray and neutron techniques

Knowledge about food structures at different length scales is key for the continued development of sustainable, tasty and healthy foods. It is critical to control, model and predict the supramolecular architecture of foods along the whole value chain: from raw materials, to their changes during processing, all the way to how products form structures during consumption and digestion. Today, advanced physical methods enable us to obtain structural information from the nanoscale- to the microscale with unprecedented resolution. The structural details can then relate to the mesoscale and microscale functionalities, important for the appeal and consumption of food products. X-ray and neutron techniques expand and strengthen the food structure characterisation toolbox. They enable in situ and in operando investigations with greater detail as well as new types of measurements that are not possible with other techniques. The knowledge gained will complement compositional and functional data obtained by other techniques, providing robustness to the interpretation of complex structural information. There are several intrinsic scientific challenges to overcome: from the lack of relevant sample environments to advanced data processing and modelling tools that consider the complexity of the food. The new frontier in food structural science can be gained through interdisciplinary collaborations not only in academia but also from the wider innovation ecosystem. This review showcases how the use of X-ray and neutron techniques is already leading to transformational knowledge in structural food science with a perspective that points to the future of this new multidisciplinary discipline. • X-ray and neutron techniques expand the food structure characterisation toolbox • in situ and in operando investigations provide transformative knowledge • X-rays and neutrons can be combined with other powerful techniques • Data handling and mathematical models must be developed • Relevant sample environments are needed