Very Active Neutral P,O-Chelated Nickel Catalysts for Ethylene Polymerization R. Soula, † J. P. Broyer, † M. F. Llauro, † A. Tomov, † R. Spitz, † J. Claverie,* ,† X. Drujon, ‡ J. Malinge, ‡ and T. Saudemont ‡ LCPP-CPE/CNRS, BP 2077, 43 Bd du 11 Nov 1918, 69616 Villeurbanne Cedex, France; and Ato-Fina, 4/8 Cours Michelet, 92800 Puteaux, France Received October 3, 2000; Revised Manuscript Received January 22, 2001 ABSTRACT: A series of highly active nickel-based neutral catalysts for ethylene polymerization is presented. These catalysts are obtained by direct complexation of simple fluorinated ketoylides onto bis- (1,5-cyclooctadiene)nickel. Catalyst formation readily occurs in the presence of an olefin, but due to the electron deficiency of the ligand, it hardly happens in the absence of an olefin or another Lewis base. Activities greater than 2 × 10 6 (gPE/gNi)/h and productivities higher than 15 × 10 6 gPE/molNi are typically observed. These catalysts are also active for the polymerization of R-olefins such as 1-hexene and 1-propene. Polymer characterization indicates that highly linear, low molecular weight PEHD is formed by these complexes. Introduction For more than 40 years, early transition metal type catalysts have generated considerable scientific interest in industry and in academia. 1-3 Despite all their ad- vantages (high activity and regio- and stereospecificity), their strongly oxophilic nature limits them to the polymerization of nonpolar monomers, with Lewis bases being a poison. Consequently, less oxophilic (softer) catalytic systems based on late transition metals have emerged, 4-6 so as to find new ways of preparing polymers or copolymers containing polar units or tol- erating functionalities. To our knowledge, the very first late transition metal catalysts for ethylene polymeri- zation are an iodorhodium complex generated from Rh- (C 2 H 4 ) 2 (acac) and I 2 reported by Kealy 7 and a dihydride tetrakistriphenylphosphine ruthenium complex dis- closed by Markham in 1972. 8 Since 1965, Yamamoto has developed a diethyl bis(2,2′-bipyridyl)iron complex for the polymerization of butadiene 9 and functionalized vinyl monomers, 10 but no activity was reported in ethylene. Considerable work has been achieved by Keim, eventually leading to the industrialization of the SHOP process. 11 Among the numerous catalysts pre- pared by Keim, 12-15 the nickel(II) complexes containing a phospho keto-ylide chelate proved to be the best candidates. In 1987, Klabunde 16 showed that, using the same kind of nickel complexes, it was possible to prepare high molecular weight polymers by removing a phos- phine with a phosphine sponge, thus opening a vacant coordination site. In addition, the co-polymerization of ethylene with an olefin bearing a polar functionality was achievable if the polar substituent was separated by a spacer of two or more methylene units. 17 Although these catalytic complexes showed a unique tolerance to polar groups, water was still described as a poison of the catalytic systems. 18 By using a phosphorus ylide as ancillary ligand, Ostoja-Starzewski was able to poly- merize ethylene without adding a phosphine sponge to the reaction medium. 19-21 These bis(ylide)nickel and palladium catalysts are very active and lead to a wide range of polyethylene molecular weight through change in the ligand. 22 After these pioneering works were disclosed, 23-25 a major contribution was achieved by Brookhart and co-workers 26,27 with the use of bulky substituted R-diimine ligands for the preparation of nickel and palladium cationic complexes. These systems proved to be very active and able to copolymerize ethylene and R-olefins with methyl acrylate. 28 Recently, neutral nickel(II) complexes based on salicylaldimine ligands were developed by Grubbs. 29,30 These complexes allow the synthesis of high molecular weight polyolefins with excellent activity. Finally, Gibson 31,32 and Ben- nett 33 reported promising work in ethylene polymeri- zation through the use of very low cost and/or low toxicity metals, such as iron and cobalt stabilized with 2,6-bis(imino)pyridyl ligands. These catalysts display excellent activities when activated with MAO. Yet, the use of MAO precludes polymerization of polar olefins or olefins in the presence of water. In a previous work, we have disclosed preliminary results about catalytic polymerization of ethylene in aqueous emulsion using binuclear phosphorus-oxygen chelating catalysts. 34 To our knowledge, for the very first time, a latex of HDPE was prepared by direct catalytic polymerization in water. 35,36 To continue our studies, we quickly realized that we needed to prepare extremely active catalysts, as we usually observed a loss of activity of 20-100 times when going from an organic medium to an aqueous dispersed medium. We chose to still work with neutral nickel catalysts as we expected MAO activation for cationic complexes to be close to impos- sible under such conditions. Moreover, we expected cationic complexes formed by borane activation, or an equivalent route, to react with hydroxide ions. Among the ligands that confer a very high activity to the catalysts, we have also decided to select those which are reasonably easy to synthesize. As it is usually accepted that activity is strongly related to the electrophilicity (acidity) of the metal, highly electrowithdrawing fluori- nated groups were incorporated onto the ligand. Here, we present first results about the synthesis of these * Corresponding author. E-mail: claverie@flamel.com. Tele- phone: (33) 4 72 78 3434. Fax: (33) 4 72 78 3435. † LCPP-CPE/CNRS. ‡ Ato-Fina. 2438 Macromolecules 2001, 34, 2438-2442 10.1021/ma001714x CCC: $20.00 © 2001 American Chemical Society Published on Web 03/16/2001