Il ruolo di Cu(0) nella polimerizzazione radicalica per trasferimento di atomo catalizzata da complessi di rame The role of Cu(0) in the atom transfer radical polymerization catalyzed by copper complexes

Atom Transfer Radical Polymerization is a powerful technique that allows building polymers with very low polydispersities and predicted molecular architectures. Recently it was born a debate about the intimate mechanism of this process when conduct in presence of metallic copper. Percec and coworkers proposed the name SET-LRP (Single Electron Transfer Living Radical Polymerization) to describe the process: they believe that Cu0 is the principal activator, CuI undergoes instantaneous and complete disproportionation, and the electron transfer (ET) follows an outher sphere path. Matyjaszewski, the inventor of ATRP, and coworkers proposed a mechanism named SARA-ATRP (Supplemental Activator Reducing Agent) in which the principal process is the classical equilibrium of ATRP, where CuI is the activator and CuII the deactivator, the ET is inner sphere and Cu0 reduces CuII to CuI and it is a supplemental activator. They studied the comproportionation reaction between CuII and Cu0, showing that this is not irrelevant and they analyzed the rapidity of activation of alkyl halides (RX), the initiators, by both CuI and Cu0, showing that Cu0 activates slower. The goal of the thesis is to enter in the debate and to definitely prove the validity of the most realistic mechanism. In this work I studied the kinetics of activation by CuI and disproportionation in DMSO and MA/DMSO 2/1 (v/v), which are the most exciting solvents considered in the debate. The complexes used have been analyzed by cyclic voltammetry, the activation and disproportionation constants were calculated by regression of data collected by chronoamperometry with RDE and simulations with Digisim program. The most studied complex was Cu/Me6TREN because it is one of the most efficient catalyst known. I also considered TPMA, PMDETA, HMTETA and bpy as ligands, and EtBrAc, MCP, BnCl and EBriB as initiators. The experiments have showed that the activation by CuI became faster in the order: bpy < HMTETA < PMDETA ≈ TPMA << Me6TREN; kact values cover the range between 10-1 to 103 M-1s-1, except by using EBriB which is very powerful and, with Me6TREN, kact is about 106 M-1s-1. The disproportionation is always very slow, kdisp is between 10-2 to 100 M-1s-1, but it is more important with PMDETA and bpy. Comparing the rapidity of these two processes it is possible to determine if CuI activates RX or it disproportionates, generating CuII and Cu0, so the latter can activate RX, as believed in SET-LRP. The ratio between the rate of activation and disporportionation depends on the ratio between the constants and the concentration of organic halides and CuI. In tipical polymerization conditions CRX >> CCuI. For every ligand analyzed vact > vdisp, and vact >> vdisp if RX = EBriB; except for bpy: with that it is necessary to use EBriB to have vact > vdisp. These results clearly show that disproportionation is slow and it can’t be instantaneous. CuI activates RX fast so it can’t rapidly generate Cu0; then, even if Cu0 would be very active, it can’t activate the halide because it is not rapidly formed in solution.