Development of a plug flow spray drying process for the manufacture of maltodextrin using a digital twin framework

This study presents the development of an integrated experimental-modelling framework for understanding and predicting the spray drying of binary solution systems through the implementation of a digital twin. Focusing on maltodextrin and lactose as materials of interest, the research aims to establish a mechanistic link between droplet-scale drying behaviour and spray drying process performance at a laboratory-scale. The work combines theoretical modelling, laboratory experimentation and data integration to provide a physically consistent description of heat and mass transfer mechanisms and their influence on product quality attributes.

A first-principles single droplet drying model (SDDM) is formulated based on a solute-fixed coordinate system to describe the coupled heat and mass transfer occurring during droplet dehydration. The governing equations are solved using a fully implicit finite difference method, which provides a numerically stable framework applicable for a wide range of drying conditions. The model predicts published data for the drying of solutions of maltodextrin and of lactose with good agreement of moisture ratio and droplet temperature evolution. The simulation successfully reproduces the characteristic drying rate behaviour and the transition from constant-rate to diffusion-controlled drying.

Further new experimental data for the drying of droplets of maltodextrin solutions are generated from existing acoustic levitation and suspended droplet apparatuses under controlled laboratory conditions. The levitator experiments at room conditions (≈24°C and 40% RH) reveal that solute concentration and acoustic voltage strongly influence droplet stability and drying rate. Increasing concentration from 10% to 50% leads to slower drying and larger final droplet diameters, while higher voltages induced deformations due to increased acoustic forces. The suspended rig experiments, performed at elevated air temperatures between 120°C and 160°C to examine the influence of temperature on droplet behaviour under typical spray drying conditions. A radiation heat transfer term was introduced in the energy balance equation of the model, but it had an insignificant effect on the overall drying behaviour. The convective heat transfer dominated at these higher temperatures and droplets with low solid content collapsed uniformly, whereas more concentrated droplets formed surface shells and internal voids.

The validated SDDM has been integrated within a plug flow spray dryer model to create a digital twin of a co-current Büchi B-290 laboratory spray dryer. A Response Surface Methodology (RSM) is applied to determine the effects of inlet drying air temperature (120–160°C), feed flow rate (6–12 mL min⁻¹) and solute concentration (30–50%) on key product properties: including outlet temperature, moisture content, water activity, yield, flowability and median particle size (D₅₀). The inlet temperature of the drying air is the dominant factor controlling moisture and water activity, while feed concentration and flow rate primarily affect yield and flowability. The plug flow model, which incorporates a Rosin–Rammler droplet size distribution at the nozzle exit, reproduces outlet air temperature within ± 3°C and predicted product moisture and particle size trends with a high accuracy.

The integration of experimental data and mathematical models establish a functional digital twin framework that links physical experiments to virtual simulations. The digital twin enabled bidirectional data exchange for model calibration and process optimisation, thereby bridging the gap between laboratory measurements and predictive modelling. The framework demonstrates its capability to forecast product attributes under new operating conditions and confirms its value as a tool that supports process-optimisation decisions. Overall, this study provides a comprehensive methodology that links the droplet-scale drying physics with process-scale modelling to offer a robust foundation for a digital twin based process development and scale up in spray drying applications within food and pharmaceutical industries.