Keywords
materials
systems
evidence synthesis
reliability
provenance
Abstract
Lithium-sulfur separator behavior emerges across interacting scales: surface chemistry and nanopores influence polysulfide transport, which shapes cell-level cycling and eventually batch-level reliability. This review examines how AI can connect those scales without replacing mechanism with correlation. Graph neural networks and attention-based sequence models are considered for representing material structure and temporal electrochemistry, while physics-informed constraints and probabilistic layers preserve known transport relationships and uncertainty. Validation is organized hierarchically, requiring success within each scale, consistency across interfaces, and external testing under changed materials, protocols, and equipment. The synthesis warns that shared trends across simulations and experiments are not proof of causal transfer. It proposes a lifecycle evidence graph linking samples, models, comparators, errors, manufacturing decisions, and operational outcomes, with named owners for each scale transition. The work is a critical review and introduces no new cell or production dataset. Its contribution is an accountable cross-scale modelling framework in which AI helps integrate evidence, but scale-up decisions remain conditioned on mechanism, provenance, calibration, and clearly stated failure consequences.
References
Debe, M. K. (2012). Electrocatalyst approaches and challenges for automotive fuel cells. Nature, 486, 43-51. https://doi.org/10.1038/nature11115
Guan, G. F., Liu, X., & Wang, L. (2026, May). A data-driven framework for contamination risk assessment in high-volume lithium-ion battery manufacturing. In 2026 6th International Conference on Machine Learning and Intelligent Systems Engineering (MLISE) (pp. 492-496). IEEE.
Hollnagel, E., Woods, D. D., & Leveson, N. (Eds.). (2006). Resilience engineering: Concepts and precepts. Ashgate.
Kingma, D. P., & Ba, J. (2015). Adam: A method for stochastic optimization. In International Conference on Learning Representations.
Kipf, T. N., & Welling, M. (2017). Semi-supervised classification with graph convolutional networks. In International Conference on Learning Representations.
Norskov, J. K., Rossmeisl, J., Logadottir, A., Lindqvist, L., Kitchin, J. R., Bligaard, T., & Jonsson, H. (2004). Origin of the overpotential for oxygen reduction at a fuel-cell cathode. The Journal of Physical Chemistry B, 108(46), 17886-17892. https://doi.org/10.1021/jp047349j
Rasmussen, C. E., & Williams, C. K. I. (2006). Gaussian processes for machine learning. MIT Press.
Tao, J., Lyu, R., & Cao, X. (2026). A scalable data governance architecture for privacy-aware intelligent learning systems in lifelong. Future-Adaptive Intelligence and Lifelong Systems, 1(1).
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., & Polosukhin, I. (2017). Attention is all you need. Advances in Neural Information Processing Systems, 30.
Wang, Z., Li, W., Wang, T., Pang, M., Kong, Z., An, J., Li, Z., Ye, J., & Xia, G. (2025). Nanoporous ZnGa2O4-modified separator as a multifunctional polysulphide barrier for advanced lithium-sulfur batteries. Electrochemistry Communications, 178, 107990. https://doi.org/10.1016/j.elecom.2025.107990