Accelerated dinuclear palladium catalyst identification through unsupervised machine learning
Julian A. Hueffel, Theresa Sperger, Ignacio Funes‐Ardoiz, Jas S. Ward, Kari Rissanen, Franziska Schoenebeck
Abstract
Learning to stabilize palladium dimers Catalyst optimization is often difficult to do rationally. Once something works, it may be unclear which specific features underpin the performance. A case in point is the stabilization of palladium(I) dimers, which has relied on a very small class of phosphine ligands. Hueffel et al . used machine learning to search for patterns in this known class of ligands and thereby guide the discovery of variants that likewise stabilize the dimers. The authors were able to synthesize eight previously unreported dimers. —JSY
Topics & Concepts
BottleneckPalladiumCluster analysisHomogeneousComputer scienceIdentification (biology)CatalysisWorkflowUnsupervised learningIn silicoArtificial intelligenceMachine learningCombinatorial chemistryChemistryMathematicsBiologyDatabaseOrganic chemistryBiochemistryEmbedded systemBotanyGeneCombinatoricsAsymmetric Hydrogenation and CatalysisMachine Learning in Materials ScienceChemical Synthesis and Analysis