Berlin 2024 – scientific programme
Parts | Days | Selection | Search | Updates | Downloads | Help
MM: Fachverband Metall- und Materialphysik
MM 3: Topical Session: Sustainable metallurgy
MM 3.2: Talk
Monday, March 18, 2024, 10:45–11:00, C 130
Data-Driven Design of Recycling Processes for Lithium Ion Batteries — •Nima Emami1, Luis Gomez-Moreno2, Anna Kemettinen2, Rodrigo Serna-Guerrero2, and Milica Todorović1 — 1University of Turku, Turku, Finland — 2Aalto Univesity, Espoo, Finland
Shifting our focus from mining to extracting materials from waste is essential for sustainable and environmental resource management. We combine lithium-ion battery recycling process simulations with data science to redesign the process for optimal materials recovery, prior to experimental validation. Starting with a model process, we altered processing parameters to simulate 10,000 process outcomes and monitored material flow to identify which parameters maximize the recovery of materials mass for graphite and LiNiMnCoO2 (NMC).
The data analysis shows that the first selected design was suboptimal: while up to 91% of graphite mass could be recovered, its purity was lacking at 70%, and NMC average mass recovery was only 6%. Given that no parameter combination could resolve this problem, we modified the processing stages and repeated the simulations with the enhanced process. We now observed the average graphene purity rise to 99%, while a much broader range of NMC mass outputs indicated that recovery of up to 92% of input mass was possible. Data analysis allowed us to determine parameter combinations that simultaneously optimise the recovery of both graphite and NMC. This study demonstrates that data-driven approaches provide new insights into recycling processes and can facilitate systemic optimisation and design.
Keywords: Lithium Ion Battery; Recycling; Data-driven Science; Process Design; Multi-Objective Optimisation