Enhanced Cycloolefin Polymerization via Dynamic Ligand Modification in Ziegler-Natta Catalysts Using Machine Learning-Driven Kinetic Control

**Abstract:** This work proposes a novel approach to enhance the polymerization of cycloolefins, specifically cyclooctene (COC), utilizing a modified Ziegler-Natta catalyst system incorporating dynamic ligand modification. We employ a machine learning (ML) algorithm trained on kinetic data to finely tune ligand exchange rates, optimizing polymerization activity and stereoselectivity. This approach overcomes limitations of traditional Ziegler-Natta catalysts in COC polymerization, namely low activity and poor control over polymer microstructure. The proposed system promises commercially viable high-performance COC polymers for advanced optical and barrier applications. Our approach leverages established catalyst chemistry and readily available ML techniques, making immediate implementation feasible.