A theoretical-data-driven coupled model for drying history and shell formation of droplets with impermeable crust: Case study on oleaster low-fibrous extract

The drying history and shell formation of droplets, particularly those forming an impermeable crust, are critical for optimizing various industrial processes. This study introduces a novel theoretical model enhanced by machine learning principles to investigate the drying of single droplets, using oleaster (Elaeagnus angustifolia L.) low-fibrous extract (OLFE) as a representative case. Experimentally informed learning-based conditions are integrated within the theoretical framework to predict key drying behaviors, including shell formation and inflation., addressing the challenges posed by high sugar content and elastic crust properties. Experimental validation demonstrated the model’s high accuracy in predicting key drying kinetics, including droplet diameter, drying time, and crust dimensions under varying conditions. Key findings reveal that higher ambient temperatures expedite drying and lead to earlier shell formation, while larger initial droplet diameters prolong drying time and result in thicker final shells. Conversely, higher initial concentrations enhance crust impermeability and strength, offering valuable insights into particle design for encapsulation and drying applications. This model bridges the gap between theoretical prediction and experimental complexity utilizing a 6-condition shell evolution model, providing a powerful tool to optimize drying processes with reduced reliance on extensive experimental trials. Its applicability extends to a wide range of materials, offering enhanced control over product quality and efficiency in spray drying and related technologies.