HTE-based opportunities for mastering olefin polymerization catalysts and processes

Chemistry in general is not an exact science. Chemical catalysis, moreover, is a purely kinetic phenomenon. This translates into the fact that discovering and even optimizing a catalyst for a desired application heavily relies on trial-and-error, and serendipitous advances are not rare. In recent years, we spent a considerable effort to extend the scope of HTE from discovery to mechanistic investigations. The general objective was to introduce protocols for ‘smart’ applications of an existing HTE workflow to complex chemical problems in polyolefin catalysis. In particular, methods for the rapid and accurate determination of the Quantitative Structure-Activity Relationship (QSAR) of representative molecular or heterogeneous catalyst formulations were implemented as the basis for statistical modeling with predictive ability. A high-performance polymerization platform (Freeslate PPR48) was integrated with state-of-art polymer characterization tools (including GPC, 13C NMR and A-CEF), and the workflow was optimized for the fast acquisition of large structure-properties databases. The approach was successfully applied to Ziegler-Natta and molecular olefin polymerization catalysts. Chain shuttling chemistry represented an extreme challenge, because operating reliably in solution at high temperature under controlled kinetic conditions in HTE scale is highly demanding; on the other hand, exploring exhaustively the complex variables hyperspace of this chemistry with conventional methods is also challenging, and requires a very long time. In the present talk, we want to share the idea that innovation in chemistry is not over even in areas that are considered mature, like e.g. catalytic olefin polymerization. Many chemical problems, including long standing ones, can rapidly find a solution as soon as adequate information becomes available. This simple and – in a way – trivial concept is often overlooked because the actual complexity of chemical systems tends to be under-estimated.