Determination of effective liquid-phase diffusion coefficients by tracer method in continuous packed columns

• Published in: Chemical engineering science 2024-08, Vol.296, p.120239, Article 120239
• Main Authors: Salmi, Tapio, Flory, Tanguy, Perez Sena, Wander, Eränen, Kari, Schmidt, Christoph, Wärnå, Johan
• Format: Article
• Language: English
• Subjects: Diffusion coefficient; Dispersion coefficient; Mathematical modelling; Porous material; Solid catalyst; Step response; Tracer
• ISSN: 0009-2509; 1873-4405
• DOI: 10.1016/j.ces.2024.120239

•In contrast to gas-phase systems, the theory and practice of determining effective diffusion coefficients in porous solid materials immersed in liquids is a challenge.•A method for the termination of effective diffusion coefficients packed beds was proposed.•Experiments were carried out with steps response technique to reveal the role of diffusion, adsorption and dispersion in packed beds.•A mathematical model was proposed and verified with experimental data. Many industrially relevant chemical processes are affected by the liquid-phase diffusion resistance in porous materials, for example in heterogeneous catalysts. Therefore, reliable experimental methods for the determination of effective diffusion coefficients in porous structures are highly desirable. In this work, a method was developed to study the internal diffusion effects in porous media by step response experiments with tracers injected to fixed beds filled with porous particles and inert material. The tracer responses were recorded at the outlet of the bed by UV–vis spectrometry. Ethanol was used as the tracer and water as the solvent. Porous aluminum oxide particles were selected as the model material. The step responses were described with a dynamic axial dispersion model, i.e. partial differential equations for the inert and catalyst sections of the bed. The partial differential equations describing the model were discretized with finite differences and the resulting ordinary differential equations were solved numerically with a solver for stiff ordinary differential equations. From the experimental data, the dispersion coefficients of the bed and the effective diffusion coefficient in the porous aluminum oxide particles were successfully determined by non-linear regression analysis. The values of the effective diffusion coefficents had a good agreement with the liquid-phase diffusion coefficients estimated from correlations. The method is applicable for any porous material.