April 7 - 11, 2025
Seattle, Washington
Symposium Supporters
2025 MRS Spring Meeting & Exhibit
BI01.03.07
We employed HSCSim simulation software to generate a comprehensive dataset through the simulation of thousands of randomly generated operational scenarios. Subsequent data analysis revealed the inherent limitations in the simulated recycling process and allowed process engineer experts to iteratively improve the process and address deficiencies. Next, we applied multi-objective optimization (MOO) methods to identify the optimal set of parameters that concurrently satisfy all established targets based on the refined process.
Our findings indicate that precise adjustments in operational parameters—such as magnetic field strength, flotation cell residence time, and the optimization of the flotation cell conditioner—can significantly enhance the separation efficiency of NMC and graphite. The optimal process attained simulated recovery rates of up to 67% for both NMC and graphite, with graphite grade levels exceeding 99%. We calculated the Pareto front and conducted MOO to derive a set of operational parameters that yield the most favorable trade-off between the recovered mass of NMC and graphite and their respective grades. This transferable data-driven approach to process engineering could bolster the circular economy in different industrial sectors by providing sustainable and efficient recycling procedures for the recovery of critical raw materials.
Data-Driven Framework to Boost Battery Recycling: Multi-Objective Optimization of Critical Material Recovery
When and Where
Apr 10, 2025
4:00pm - 4:15pm
4:00pm - 4:15pm
Summit, Level 3, Room 325
Presenter(s)
Co-Author(s)
Milica Todorovic1,Nima Emami1,Luis Gomez-Moreno2,Anna Klemettinen2,Rodrigo Serna-Guerrero2
University of Turku1,Aalto University2
Abstract
Milica Todorovic1,Nima Emami1,Luis Gomez-Moreno2,Anna Klemettinen2,Rodrigo Serna-Guerrero2
University of Turku1,Aalto University2
The electrification drive has markedly increased the demand for raw materials necessary for battery production, prompting the need for superior recycling practices that can enhance the recovery of critical materials from end-of-life electrical waste. In this study, we propose a novel data-driven framework to optimize materials recovery from battery waste in the recycling process. We performed high-throughput simulations of the mechanical separation stage of the lithium-ion battery recycling, focusing on the mass flow of nickel-manganese-cobalt oxide (NMC) and graphite. The primary objective was to identify the operational parameters of a feasible mechanical separation process that maximize the mass recovery and grade (purity) of graphite and NMC.We employed HSCSim simulation software to generate a comprehensive dataset through the simulation of thousands of randomly generated operational scenarios. Subsequent data analysis revealed the inherent limitations in the simulated recycling process and allowed process engineer experts to iteratively improve the process and address deficiencies. Next, we applied multi-objective optimization (MOO) methods to identify the optimal set of parameters that concurrently satisfy all established targets based on the refined process.
Our findings indicate that precise adjustments in operational parameters—such as magnetic field strength, flotation cell residence time, and the optimization of the flotation cell conditioner—can significantly enhance the separation efficiency of NMC and graphite. The optimal process attained simulated recovery rates of up to 67% for both NMC and graphite, with graphite grade levels exceeding 99%. We calculated the Pareto front and conducted MOO to derive a set of operational parameters that yield the most favorable trade-off between the recovered mass of NMC and graphite and their respective grades. This transferable data-driven approach to process engineering could bolster the circular economy in different industrial sectors by providing sustainable and efficient recycling procedures for the recovery of critical raw materials.
Symposium Organizers
Jurgen Eckert, Montanuniversität Leoben
Ryan Ott, Ames Laboratory
Min-Ha Lee, Korea Institute of Industrial Technology
Alessa Hool, ESM Foundation
Symposium Support
Bronze
Korea Institute of Materials Science
Korea Institute of Materials Science
Session Chairs
Ryan Ott
Mihai Stoica
In this Session
