Direct design comparison of Li-ion cells in the formats PHEV1, pouch, 18650, and 21700 built on pilot-scale

Summary:

Commercial Li-ion cells are produced in a variety of different formats, such as cylindrical (e.g. 18650, 21700, 46800), as well as different pouch and prismatic formats. In scientific literature, there is a lack of knowledge on the effects on cell design on performance and aging. Due to differences in chemistry, it is difficult to investigate this topic with commercial cells [1]. Ideally, knowledge on cell designs could be built up by preparing cells of different formats with the same electrodes, electrolytes, and separators. This is the topic of the present talk where the industrially-relevant cell formats PHEV1, multi-layer pouch, 18650, and 21700 [2-4] are directly compared by experiments. For this purpose, all cells were built reproducibly at ZSW’s pilot-lines. The data are put into context of commercial cells and reveal general trends.

Comparison of cylindrical cell formats:
A direct experimental comparison of 18650 and 21700 cells shows that the energy content increases by a factor of 1.5 [2], which is consistent with calculations using cylinder volumes [1] and Archimedean spirals [2]. We estimate that the energy content further increases from the 18650 to Tesla’s 46800 format by a factor of ~8.4. With increased size, less cells have to be produced per Wh, leading most likely to cost reduction.
While cells in the 18650 and 21700 formats with the same chemistry show a very similar capacity fade over 1000 cycles (1C / 25°C) [2], we observe an improvement of cycle life by ~14% for 21700 with tabs made from the current collecting foils compared to conventionally welded tabs [4].

Comparison of 21700, pouch, and PHEV1 formats:
The main differences between these formats are their capacities, sizes, as well as tab and the jellyroll/stack configurations (flat-wound, stacked, wound). We report the following trends:
• Low C-rates (≤ 1C) lead to very similar voltage curves for all formats
• Higher C-rates lead to differences in the voltage curves, especially for cells with a lower ratio of cell volume to cell surface.
• The cell’s volume/surface ratio determines the influence of the C-rate on the maximum temperature Tmax measured for discharge of fully charged cells with a constant current.
• The cell impedance (1 kHz) is influenced mainly by the electrode area and the tab configuration. The PHEV cells with a tabless design and the largest electrode area therefore show the lowest impedance of only ~1 mΩ.

[1] J.B. Quinn et al., J. Electrochem. Soc. 165 (2018) A3284–A3291. https://doi.org/10.1149/2.0281814jes
[2] T. Waldmann et al., J. Power Sources. 472 (2020) 228614. https://doi.org/10.1016/j.jpowsour.2020.228614
[3] T. Waldmann et al., J. Electrochem. Soc. 168 (2021) 090519, https://doi.org/10.1149/1945-7111/ac208c
[4] T. Waldmann et al., Processes 9 (2021) 1908, https://doi.org/10.3390/pr9111908

Acknowledgment:
Funding of the projects E-QUAL (03XP0252A) and RollBatt (03XP0245A) by the German Federal Ministry of Education and Research (BMBF) are gratefully acknowledged.