Final Report Summary - METMED-CRP (Toward new polymeric materials by metal-mediated controlled radical polymerization)
Vinyl acetate (VAc) is an important monomer because Poly(vinyl acetate) (PVAc) and its hydrolysed product, Poly(vinyl alcohol) (PVA), have numerous applications in coating, fibre, textile, adhesive, pharmaceutical and photographic industries. VAc can only undergo radical polymerisation. Therefore, great efforts have been made to explore the living radical polymerisation of VAc for making PVAc and PVA with controlled molecular weights and structures. Organometallic mediated radical polymerisation (OMRP) has become a preferred tool in controlling the polymerisation of less reactive monomers such as VAc. OMRP rests on the reversible deactivation of growing radical chains with formation of a direct metal-carbon bond with a transition metal complex. Reversible binding of very reactive radicals requires metals that do not form strong metal-carbon bonds and therefore iron(II) and copper(I) were considered promising candidates. There is no example for OMRP of VAc with Cu-based or Fe-based catalysts. In addition, Fe-based catalysts have attracted particular attention in Atom transfer radical polymerisation (ATRP) owing to their low toxicity and low cost. Given the application value of iron in polymerisation, further development of OMRP using copper or iron as trapping agent seems to be significant.
For Cu(I), we prepared adducts of complexes TpCu and Tp*Cu (Tp = tris(pyrazolyl)borate; Tp* = tris(3,5-dimethyl-pyrazolyl)borate) with VAc, but this olefin is too weakly binding to yield stable adducts. Nevertheless, a preliminary study of the VAc polymerisation under OMRP conditions (thermal initiation by a diazo compound) shows that whereas no radical trapping is insured by Tp*Cu, efficient trapping and reversible radical release occurs for TpCu. These results are promising but so far insufficient to establish the OMRP principle on the basis of copper and further work is needed.
Meanwhile, we had better success using an iron system, Fe(acac)2 (acac = acetyl acetonate). With this compound, the VAc polymerisation is controlled, albeit poorly. The molecular weights (Mn) of the resulting polymer are much higher than the theory. However, the resulting polymer had relatively low polydispersities (PDI = 1.29-1.36) and the Mn increased proportionally with the conversion. Certain additives, particularly phosphines, slow down the polymerisation and have a positive effect on the controllability. Dimethylphenylphosphine (DMPP) gave the best results (smaller discrepancy between observed and calculated Mn; PDI = 1.16-1.21). This indicates that the PVAc chains are reversibly released from the (acac)2Fe-PVAc dormant species, as expected for an OMRP mechanism. We have also isolated a metal-capped short oligomer, (acac)2Fe-(VAc)n-R0 (R0 is the primary radical of the initiator), fully characterised it by NMR, EPR, and chemical derivatisation (demetallation with thiols), and proved the reversible radical release from it, by using it as a thermal initiator for the OMRP of VAc.
We have also carried out a thorough 1H NMR study of Fe(acac)2 solutions in a wide variety of non-coordinating and coordinating solvents, as well as its interaction with Et3N, pyridine, PMe2Ph, and R2PCH2CH2PR2 [R = Ph (dppe), Et (depe)] in C6D6. The study reveals that the commercial compound is contaminated by important amounts of Fe(acac)3. Still by using 1H NMR, we could investigate the reversible equilibria between Fe(acac)2 and neutral donor molecules (L), leading to a variety of Fe(acac)2(L)x adducts (x = 1/2, 1, 2) that are relevant to insure the controllability of the radical polymerisation process.
For Cu(I), we prepared adducts of complexes TpCu and Tp*Cu (Tp = tris(pyrazolyl)borate; Tp* = tris(3,5-dimethyl-pyrazolyl)borate) with VAc, but this olefin is too weakly binding to yield stable adducts. Nevertheless, a preliminary study of the VAc polymerisation under OMRP conditions (thermal initiation by a diazo compound) shows that whereas no radical trapping is insured by Tp*Cu, efficient trapping and reversible radical release occurs for TpCu. These results are promising but so far insufficient to establish the OMRP principle on the basis of copper and further work is needed.
Meanwhile, we had better success using an iron system, Fe(acac)2 (acac = acetyl acetonate). With this compound, the VAc polymerisation is controlled, albeit poorly. The molecular weights (Mn) of the resulting polymer are much higher than the theory. However, the resulting polymer had relatively low polydispersities (PDI = 1.29-1.36) and the Mn increased proportionally with the conversion. Certain additives, particularly phosphines, slow down the polymerisation and have a positive effect on the controllability. Dimethylphenylphosphine (DMPP) gave the best results (smaller discrepancy between observed and calculated Mn; PDI = 1.16-1.21). This indicates that the PVAc chains are reversibly released from the (acac)2Fe-PVAc dormant species, as expected for an OMRP mechanism. We have also isolated a metal-capped short oligomer, (acac)2Fe-(VAc)n-R0 (R0 is the primary radical of the initiator), fully characterised it by NMR, EPR, and chemical derivatisation (demetallation with thiols), and proved the reversible radical release from it, by using it as a thermal initiator for the OMRP of VAc.
We have also carried out a thorough 1H NMR study of Fe(acac)2 solutions in a wide variety of non-coordinating and coordinating solvents, as well as its interaction with Et3N, pyridine, PMe2Ph, and R2PCH2CH2PR2 [R = Ph (dppe), Et (depe)] in C6D6. The study reveals that the commercial compound is contaminated by important amounts of Fe(acac)3. Still by using 1H NMR, we could investigate the reversible equilibria between Fe(acac)2 and neutral donor molecules (L), leading to a variety of Fe(acac)2(L)x adducts (x = 1/2, 1, 2) that are relevant to insure the controllability of the radical polymerisation process.