Reducing the cost of redox flow batteries (RFBs) is critical to achieving broad commercial deployment of large-scale energy storage systems. This can be addressed in a variety of ways, such as reducing component costs or improving electrode design. The aim of this work is to better understand the relationship between electrode microstructure and performance. Four different commercially available carbon electrodes were examined – two cloths and two papers (from AvCarb® and Freudenberg Performance Materials) – and a comprehensive study of the different pore-scale and mass-transport processes is presented to elucidate their effect on the overall cell performance. Electrochemical measurements were carried out in a non-aqueous organic flow-through RFB with these different electrodes, using two supporting solvents (propylene carbonate and acetonitrile) and at a variety of flow rates. Electrode samples were scanned using X-ray computed tomography, and a customised segmentation technique was employed to extract several microstructural parameters. A pore network model was used to calculate the pressure drops and permeabilities, which were found to be within 1.26 × 10−11 and 1.65 × 10−11 m2 for the papers and between 8.61 × 10−11 and 10.6 × 10−11 m2 for the cloths. A one-dimensional model was developed and fit to polarisation measurements to obtain mass-transfer coefficients, km, which were found to be between 1.01 × 10−6 and 5.97 × 10−4 m s−1 with a subsequent discussion on Reynolds and Sherwood number correlations. This work suggests that, for these fibrous materials, permeability correlates best with electrochemical cell performance. Consequently, the carbon cloths with the highest permeability and highest mass-transfer coefficients, displayed better performances.
Bibliographical noteFunding Information:
The authors would like to acknowledge Dr Vladimir Yufit for initial help in applying for this International Collaboration Seed Funding. We would like to thank Dr Aayan Banerjee for his insightful discussions in developing the model. BAS acknowledges Dr Peter A.A. Klusener and Dr Nicola Menegazzo (Shell) and thanks the EPSRC and Shell Global Solutions International B.V. for financial support. AGL acknowledges CONACYT-SENER for financial support. CPM acknowledges BECAS CHILE-CONICYT, Ministry of Education Chile Scholarship and EPSRC UK grant EP/L014289/1 “Lower Cost and Longer Life Flow Batteries for Grid Scale Energy Storage” for financial support. KMT, KVG, AFC and FRB acknowledge financial support from the Joint Center for Energy Storage Research (De-AC02-06CH11357). KMT and KVG acknowledge additional funding from the NSF Graduate Research Fellowship (1122374). Any opinion, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation. A.F.C. acknowledges the Swiss National Science Foundation for funding his postdoctoral fellowship (Grant No. P2EZP2_172183) and from the Dutch Science Foundation under the Veni Award (#17324). We would also like to thank Mr. Bertrand J. Neyhouse for assistance in the electrolyte preparation.
- 1D model
- Carbon electrodes
- Non-aqueous redox flow batteries
- XCT imaging