Shape factor optimisation for the distribution of relaxation times to better deconvolute electrochemical impedance spectra

• Published in: Journal of electroanalytical chemistry (Lausanne, Switzerland) Switzerland), 2024-06, Vol.962, p.118272, Article 118272
• Main Authors: Wang, Jia, Huang, Qiu-An, Wang, Juan, Zhang, Jiujun
• Format: Article
• Language: English
• Subjects: Collaborative optimization strategy; Distribution of relaxation times; Penalty factor; Shape factor; Tikhonov regularization
• ISSN: 1572-6657; 1873-2569
• DOI: 10.1016/j.jelechem.2024.118272

•Revealed the mechanism and influence of the shape factor on DRT deconvolution;•Evaluated the optimisation methods of the shape factor by the synthetic EIS data;•Discussed the coupling effect of the shape and penalty factors on DRT deconvolution;•Proposed a collaborative optimisation strategy to improve the accuracy of DRT result; Electrochemical impedance spectroscopy (EIS) is a powerful diagnosis tool for the performance of electrochemical energy storage and conversion devices. One challenge for EIS-based diagnosis is the difficulty in separating highly overlapped impedance spectra. To overcoming this challenge, the distribution of relaxation times (DRT) method based on Tikhonov regularization algorithm can be employed. However, the accuracy and stability of the DRT deconvolution method not only depend on the penalty factor (the actual regularization tool), but also on the rarely studied shape factor (a potential regularization tool). In this study, a comprehensive investigation on the effect of the shape factor on both the accuracy and the stability of the DRT deconvolution method was conducted. First, the influence of the shape factor on the DRT deconvolution method was theoretically derived and numerically simulated. Second, the optimization methods for the shape factor and the selection of regularization matrices for DRT deconvolution were quantitatively evaluated using synthetic impedance data. Third, the coupling effect of the shape and penalty factors on the DRT deconvolution method was analyzed quantitatively. Finally, the collaborative optimization strategy was proposed and evaluated. The method presented in this paper can be used to improve the accuracy and reliability of the DRT to decode highly overlapped impedance spectra.