Quantitative Analysis of Tomographic Images for Understanding Discharge Phenomena in Alkaline Zn-MnO₂ Batteries

<i>In situ</i> synchrotron CT was used to obtain volumetric reconstructions of sealed commercial Zn-MnO₂ batteries. The voxel resolution was 2.93 <i>μ</i>m/voxel with sufficient field of view (FOV) that a volume of approximately 200 mm³ encompassing the entire battery diameter could be probed in a reasonable time frame. The reconstructions show the internals of partially discharged anodes, which visualizes the morphology and distribution of the active phases. A novel segmentation algorithm converted the tomographic images to the discrete phases present, allowing quantitative analyses to be performed. In particular, one dimensional (1D) radial profiles of ZnO and undischarged Zn were calculated, for direct comparison to the output of a computational battery model. Within these batteries, the Zn anode includes a current collecting pin that is frequently off centered. A pseudo-cylindrical coordinate system was designed to characterize the competing ionic and electronic effects in these batteries. This allows cylindrical symmetry to be maintained, despite the off centered current collecting pin. This is an imperative step in directly comparing quantitative experimental results to 1D model predictions.<br/><br/>The discharge of an alkaline Zn-MnO₂ battery involves a dissolution-precipitation mechanism at the Zn anode, where Zn dissolves into the electrolyte and precipitates out as ZnO. The zincate ion is highly mobile in alkaline media, leading to active material redistribution and complex morphologies of ZnO discharge products. Throughout the discharge, a Zn particle must be connected to the electronically conducting Zn network, while also maintaining contact with the electrolyte to sustain further discharge.¹ Localized precipitation of ZnO will impede hydroxide transport through that region and degrade cell performance.² Modeling has been used to simulate the performance of these commercial batteries, though model predictions do not match experimental results under numerous discharge conditions that are typical in real-world use cases.³ By characterizing commercial Zn-MnO₂ batteries under various discharge conditions, this work seeks to improve modeling capabilities through a better understanding of the discharge behavior of Zn anodes.⁴<br/><br/>These quantitative analyses showed Zn and ZnO distribution was highly dependent on the discharge protocol. A pulsed discharge allows more microporous and spatially distributed ZnO to form off of the active Zn surface, reducing the passivating effects of ZnO formation.^4,5 Furthermore, the rate at which the cell is discharged dictates where the bulk of the reaction occurs in the anode. At high rates, a conventional reaction zone near the separator can be seen and is verified by model predictions. However, at lower rates this reaction zone inverts and has the majority of Zn dissolution and ZnO precipitation near the current collecting pin, which is not predicted by current models.³ By analyzing the discharge of Zn anodes using segmented tomography in pseudo-cylindrical coordinates, the factors that dictate ZnO distribution and morphology can be more thoroughly understood and used to guide the development of battery models.<br/><br/><br/><b>Acknowledgements</b><br/>This research was supported by funding from Energizer Holdings, Inc. We acknowledge collaborators Xiaotong Chadderdon, Matthew Wendling, Andrew Chihpin Chuang, and John Okasinski. This research also used resources of the Advanced Photon Source beamline 6-BM, a U.S. Department of Energy (DOE) Office of Science User Facility operated for the DOE Office of Science by Argonne National Laboratory under Contract No. DE-AC02-06CH11357.<br/><br/><b>References</b><br/>[1] I. Arise et al 2013 J. Electrochem. Soc. 160 D66<br/>[2] Q. C. Horn and Y. Shao-Horn 2003 J. Electrochem. Soc. 150 A652<br/>[3] E. J. Podlaha and H. Y. Cheh 1994 J. Electrochem. Soc. 141 15<br/>[4] D. P. Guida et al 2023 J. Power Sources 556 232460<br/>[5] C.G. Smith 1978 Lawrence Berkeley National Laboratory