Micro packed-bed reactors (μPBRs) offer advantages such as high specific surface area, enhanced heat and mass transfer, and improved reaction efficiency for various chemical processes. However, the significant pressure drop is a critical challenge in these systems. Predicting the pressure drop is essential for practical applications, as it determines the required pumping power and overall system efficiency. Obtaining three-dimensional (3D) real-space coordinates of the packed particles is crucial for investigating the relationship between design parameters (
i.e., particle size, shape, material, and packing method) and their performance such as pressure drop, mass transfer, heat transfer, and reaction efficiency. This study presents an innovative approach for characterizing μPBRs by visualizing their 3D packing structure of fluorescent-labelled particles (< 2 μm in diameter) using confocal microscopy. The pressure drop of the packed beds is predicted through finite-element calculations based on these 3D coordinates, providing insight into the relationship between the packing structure and pressure drop.
Fluorescent-labelled silica particles with different diameters (1.71 μm and 1.35 μm) were utilized in packed beds that could be visualized through confocal microscopy. Pressure drop values of the packed beds were measured. The values were different from theoretical predictions (Kozeny-Carman equation), indicating that the pressure drop of colloidal particles cannot be accurately expressed using conventional theories. To validate the relationship between the packing structure and pressure drop, we implemented 3D computational fluid dynamics (CFD) using the finite-element method. The experimentally obtained particle coordinates were directly used as input in COMSOL Multiphysics® to create accurate digital representations of the packed beds. CFD simulations through these structures reproduced the deviations from theoretical predictions which were found in the experiments. We concluded that the visualization method provided us sufficient structural information of packed beds with particles smaller than 2 μm for accurate prediction of pressure drop.
