Experimental analysis of lithium-ion cell procurement: Quality differences, correlations, and importance of cell characterization

Abstract

Since every production process exhibits slight variances, lithium-ion cells are usually tested for their function and quality by the manufacturer during end-of-line testing, whereby specific tolerance limits are generally not communicated to end users. In addition, commercial cells are usually sold through various distribution channels and intermediaries, creating a black box regarding the actual cell properties delivered, unless a clear requirement specification supplier commitment has been defined. In this work, 250 cylindrical 18650 cells from the same manufacturer and model were obtained via two distribution channels and quality grading categories and extensively characterized using established methods of cell characterization. The relationship between the various measurements and cell characteristics was evaluated and visualized to reveal interrelated factors which remain hidden in aggregate observations. The results show that the choice of vendor can have a significant impact on the performance characteristics obtained for the supposedly same product, with strong scatter being evident both between and within the three batches. In addition, the given grading classes do not correlate with the measured cell performance in terms of capacity, internal resistance, cycling performance or a combination of the aforementioned criteria. This could be caused by an imperfect classification by the manufacturer or a mixing of quality classes at the vendor. Hence, the investigation encourages the implementation of appropriate cell quality categorization depending on the application and a profound exchange of data between vendors and customers, in order to prevent the utilization of under-performing cells.