Numerical Investigation of 2D Ordered Pillar Array Columns: An Algorithm of Unit-Cell Automatic Generation and the Corresponding CFD Simulation

• Published in: Separations Vol. 12; no. 7; p. 184
• Main Authors: Jiang, Qihao, Rocca, Stefano, Shaikhuzzaman, Kareem, Dimartino, Simone
• Format: Journal Article
• Language: English
• Published: Basel MDPI AG 01.07.2025
• Subjects: Algorithms; Arrays; CFD; Chromatography; Computational fluid dynamics; Constraints; Fluid dynamics; fluid velocity profiles; Homogeneity; Liquid chromatography; Morphology; morphology generation; ordered pillar array columns; Particle shape; Performance evaluation; Permeability; Porosity; Separation; separation performance; Symmetry; Tracers (Chemistry); Unit cell; Velocity distribution; Viscosity
• ISSN: 2297-8739; 2297-8739
• DOI: 10.3390/separations12070184

This paper presents a numerical investigation into the generation of 2D ordered pillar array columns for liquid chromatography columns, focusing on the development of an algorithm for the automatic creation of unit-cell morphologies and their subsequent computational fluid dynamics (CFD) simulation. The algorithm is developed to incorporate functional and operational constraints, which ensure that the generated structures are permeable and suitable for chromatographic separations. The functional constraints include the principal pathway and no dry void constraints, while the operational constraints involve symmetry and porosity thresholds. The algorithm’s efficacy is demonstrated with a reduction rate of 97.8% for order 5 matrices. CFD simulations of the generated morphologies reveal that the homogeneity of the fluid velocity profile within the unit cell is a key determinant of separation performance, suggesting that refining the resolution of discrete unit cells could enhance separation efficiency. Future work will explore the inclusion of more complex morphologies and the impact of particle shape and size on separation efficiency.