Jack Ledingham , Jonathan R. Howse , Alasdair N. Campbell , Ben in ‘t Veen , Lucas Keyzer , Kyra L. Sedransk Campbell
{"title":"Experimental validation of the presence of flow maldistribution in electrodialysis stacks and its effect on the limiting current density","authors":"Jack Ledingham , Jonathan R. Howse , Alasdair N. Campbell , Ben in ‘t Veen , Lucas Keyzer , Kyra L. Sedransk Campbell","doi":"10.1016/j.memsci.2024.123494","DOIUrl":null,"url":null,"abstract":"<div><div>Flow maldistribution between channels in electrodialysis has previously been simulated and is predicted to have a substantial impact on the limiting current density (LCD) and thus constitutes a barrier to industrial implementation. Using computational fluid dynamics (CFD) simulations and a 1-D circuit-based model, it was previously concluded that maldistribution is both prevalent in electrodialysis and adversely affects operation through a reduction of the LCD. In this work, the presence and impact of maldistribution are confirmed experimentally using red-blue particle image velocimetry (PIV) and through measuring the LCD as a function of the degree of maldistribution, respectively. In the PIV experiments, 50 μm particles were suspended and were flowed through a glass flow cell with the same geometry as those used in the CFD simulations. These spheres were imaged using a single exposure photograph and two pulsed LED bursts and their velocities subsequently calculated. The imaging of many particles demonstrated significant maldistribution in line with what had been previously predicted. The effect of maldistribution on operation was experimentally validated by measuring the LCD of an electrodialysis stack as a function of the degree of maldistribution. The degree of maldistribution was independently controlled by varying the flow rate and the number of cell pairs while accounting for confounding effects. The measured LCD was found to decrease with an increasing degree of maldistribution. A further CFD study was performed, comparing the degree of maldistribution in U and Z configuration stacks. It was demonstrated that while at low flow rates, maldistribution is worse for U-type geometries, at high flow rates, it is worse for Z-type geometries. Overall, the results presented in this work conclusively demonstrate that maldistribution exists within electrodialysis and significantly affects the LCD. This consequently validates previous modelling results and highlights maldistribution as an important phenomenon in electrodialysis that should be considered when optimisation is performed.</div></div>","PeriodicalId":368,"journal":{"name":"Journal of Membrane Science","volume":"716 ","pages":"Article 123494"},"PeriodicalIF":8.4000,"publicationDate":"2024-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Membrane Science","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0376738824010883","RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENGINEERING, CHEMICAL","Score":null,"Total":0}
引用次数: 0
Abstract
Flow maldistribution between channels in electrodialysis has previously been simulated and is predicted to have a substantial impact on the limiting current density (LCD) and thus constitutes a barrier to industrial implementation. Using computational fluid dynamics (CFD) simulations and a 1-D circuit-based model, it was previously concluded that maldistribution is both prevalent in electrodialysis and adversely affects operation through a reduction of the LCD. In this work, the presence and impact of maldistribution are confirmed experimentally using red-blue particle image velocimetry (PIV) and through measuring the LCD as a function of the degree of maldistribution, respectively. In the PIV experiments, 50 μm particles were suspended and were flowed through a glass flow cell with the same geometry as those used in the CFD simulations. These spheres were imaged using a single exposure photograph and two pulsed LED bursts and their velocities subsequently calculated. The imaging of many particles demonstrated significant maldistribution in line with what had been previously predicted. The effect of maldistribution on operation was experimentally validated by measuring the LCD of an electrodialysis stack as a function of the degree of maldistribution. The degree of maldistribution was independently controlled by varying the flow rate and the number of cell pairs while accounting for confounding effects. The measured LCD was found to decrease with an increasing degree of maldistribution. A further CFD study was performed, comparing the degree of maldistribution in U and Z configuration stacks. It was demonstrated that while at low flow rates, maldistribution is worse for U-type geometries, at high flow rates, it is worse for Z-type geometries. Overall, the results presented in this work conclusively demonstrate that maldistribution exists within electrodialysis and significantly affects the LCD. This consequently validates previous modelling results and highlights maldistribution as an important phenomenon in electrodialysis that should be considered when optimisation is performed.
期刊介绍:
The Journal of Membrane Science is a publication that focuses on membrane systems and is aimed at academic and industrial chemists, chemical engineers, materials scientists, and membranologists. It publishes original research and reviews on various aspects of membrane transport, membrane formation/structure, fouling, module/process design, and processes/applications. The journal primarily focuses on the structure, function, and performance of non-biological membranes but also includes papers that relate to biological membranes. The Journal of Membrane Science publishes Full Text Papers, State-of-the-Art Reviews, Letters to the Editor, and Perspectives.