Machine Learning-Guided Development of Trialkylphosphine Ni^(I) Dimers and Applications in Site-Selective Catalysis

Owingto the unknown correlation of a metal's ligand andits resulting preferred speciation in terms of oxidation state, geometry,and nuclearity, a rational design of multinuclear catalysts remainschallenging. With the goal to accelerate the identification of suitable ligands that form trialkylphosphine-deriveddihalogen-bridged Ni-(I) dimers, we herein employed an assumption-basedmachine learning approach. The workflow offers guidance in ligandspace for a desired speciation without (or only minimal) prior experimentaldata points. We experimentally verified the predictions and synthesizednumerous novel Ni-(I) dimers as well as explored their potentialin catalysis. We demonstrate C-I selective arylations of polyhalogenatedarenes bearing competing C-Br and C-Cl sites in under5 min at room temperature using 0.2 mol % of the newly developed dimer,[Ni-(I)(& mu;-Br)PAd(2)(n-Bu)](2), which is so far unmet with alternative dinuclear or mononuclearNi or Pd catalysts.