Toward Rigorous Validation of Li-S Battery Models

Abstract

Achieving Li-S batteries’ promise of significantly higher gravimetric energy density and lower cost than Li-ion batteries requires researchers to delineate the most important factors affecting the performance of this technology. By encoding this knowledge into a mathematical model, understanding is made precise, quantitative, and predictive. However, the complex and unknown mechanisms of Li-S batteries have multiple proposed models with relatively few informative quantitative comparisons to experimental data. Without further testing, many proposed models do not have enough evidence to claim predictive power. The conclusions drawn from these models regarding the internal dynamics of Li-S cells may be correct, but the lack of evidence provided leaves these conclusions uncertain. Consequently, a minimum set of testing procedures for model validation is proposed. Moreover, in the absence of an accepted standard model, a novel zero dimensional model is proposed in this work. The model improves upon several existing models while remaining as simple as possible. The model is quantitatively predictive, as demonstrated by out-of-sample predictions of experimental discharge resistance. Finally, this model and others have been implemented using PyBaMM. Therefore, the open access code allows rapid modifications of this model by all researchers.