(129a) Improved Understanding of Materials’ Formulations and Design By Combining Multiscale Molecular Modeling with AI/ML Techniques

Data-driven modeling approaches continue to get more attention both in academia and industry. This is evident from the tsunami of papers published in the last decade on data driven modeling, including AI/ML techniques, and the success it actually has in industry when it comes to model accuracy and prediction, and the availability of relatively cheap computing power, while at the same time allowing companies to start making use of their large internal data pools.

In this work we discuss four industrial use cases. The first one deals with battery manufacturing where we simulate the processes to generate anode and cathode materials from slurry drying to calandering and we will showcase how to integrate these results in battery cell testing. [1]. The second case discuss the design of the battery separator material, which are typically polymer blends that experience non-llnear stress/strain behavior in typically small temperature range. We apply coarse-grained simulations techniques to be able to address the required time and length scales and show how to obtain the optimum polymer blend and corresponding material safetely specifications. Thirdly, we investigated by molecular simulation how to boost the performance of lithium ionic conductivity of solid state materials [2]. And finally, we showcase how multiscale modeling can be applied to the optimization of complex nutrient formulations for medical applications and how to relate it to key product characteristics like viscosity and texture.

When applicable we will apply AI/ML techniuqes to further enhance the robustness of the modeling in terms of predictive power and raw speed, where have used both in-house available methods and software tools from Google and Citrin Informatics [3].

[1] Microscopic investigation of the slurry drying process and binder migration in Li-ion battery anodes using multiscale-simulation methods, S Indrakumar, J Breitenbach, S Rudolf, JW Handgraaf, L Riewel, New Materials for Future Mobility 883 (Ingénieurs de l'Automobile), 35-40.

[2] Adeli et al., Boosting Solid-State Diffusivity and Conductivity in Lithium, Superionic Argyrodites by Halide SubstitutionAngew. Chem. 2019, 131, 8773 –8778.

[3] https://citrine.io/wp-content/uploads/2021/04/Case-Study-Panasonic-Techn...