Catalyst Design DOI: 10.1002/ange.201003465 Towards Selective Ethylene Tetramerization** Sebastiano Licciulli, Indira Thapa, Khalid Albahily, Ilia Korobkov, Sandro Gambarotta,* Robbert Duchateau,* Reynald Chevalier, and Katrin Schuhen The metallacycle mechanism formulated to account for the excellent selectivity of catalytic ethylene-trimerization pro- cesses is today well-established. [1–3] Further ring expansion through the insertion of a fourth ethylene unit is believed to afford 1-octene. [4] However, selectivity becomes a major challenge in this event. If a fourth molecule of ethylene can be inserted readily into a seven-membered ring, it is clearly very difficult to prevent further expansion of the nine-membered ring (and thus the formation of heavier oligomers). In fact, in the entire patent and academic literature, only two homoge- neous catalytic systems, discovered by researchers at Sasol [4a] and SK Energy, [4b] have been described that are capable of producing a substantial excess of 1-octene (about 70 %) over other a-olefins. The search for a catalytic system capable of producing 1-octene as the only product is still being pursued actively. One may even question whether selective tetrame- rization to produce solely 1-octene may ever be possible, unless an alternate mechanistic pathway is followed. In our search for highly selective ethylene-tetramerization catalysts, we selected the 2,2’-dipyridylamine ligand with an alkylated central nitrogen atom. Alkylation of the central N atom to prevent anionization was deemed necessary to maintain the possibility of cationizing the monovalent metal center, as necessary for catalytic activity. [1p] Alkylation of the central nitrogen atom was also expected to diminish the established tendency of the ligand to form multiply bonded dimers or higher aggregates, [5] because of sterically induced deformation of the ligand backbone. [6] A range of substituted (2-C 5 H 4 N) 2 NR derivatives and the corresponding trivalent chromium complexes [{(2- C 5 H 4 N) 2 NR}CrCl 3 (thf)] (1a : R = Me; 1b : R = CH 2 CMe 3 ; 1c :R = CH 2 SiMe 3 ; 1d :R = C 16 H 33 ; 1e :R = benzyl; 1f :R = C 3 H 6 Si(OEt) 3 ; 1g : R = C 4 H 8 OEt) were synthesized readily. The monomethyl Cr III derivative [{(2- C 5 H 4 N) 2 NCH 2 SiMe 3 }CrMeCl 2 (thf)] (2c) was prepared by treating 1c with a stoichiometric amount of MeLi in THF or by the direct treatment of [CrMeCl 2 (thf) 3 ] with the ligand. The structures were all very similar (Figure 1). When activated with methylaluminoxane (MAO) at 50 8C, all trivalent complexes underwent a vigorously exothermic reaction after an induction period of several minutes to form large amounts of a heavy a-olefin. In all cases, the 13 C NMR spectrum showed the presence of a vinylic residue and a complete lack of branching: features indicative of linear a- olefins. Under isoparabolic conditions, with temperatures rising to a maximum of 110 8C, we observed that, aside from polyethylene (PE) wax, a sizeable amount of highly pure 1- octene was formed, as shown by NMR spectroscopy and GC– MS (Figure 2). The thermal behavior of the reaction was remarkable. After an induction period of about 4 min, the temperature increased very rapidly to reach about 110 8C. When the reaction temperature was maintained constant at 80 8C with the aid of a cooling coil in the interior of the reactor, the waxy a-olefin appeared to be the sole product of the reaction (only traces of 1-octene were detected; Table 1, entry 7). A lower catalyst loading and variable pressure did not affect the activity significantly. [7] Attempts to diminish the amount of PE wax by carrying out the catalytic reaction in the presence of hydrogen gas did not affect the outcome. MAO appears to be the only usable activator, since no catalytic activity was observed with other common alkyl aluminum compounds, including trimethylaluminum (TMA), triiso- butylaluminum, tetraisobutyldialuminoxane, triethyl- aluminum, and diethylaluminum chloride. [8] Interestingly, even the use of TMA-depleted MAO gave no reaction. Conversely, when a small amount of TMA (10%) was added Figure 1. Thermal-ellipsoid (50 % probability) plots of 1b, 1g, and 2c. [11] [*] Dr. S. Licciulli, I. Thapa, K. Albahily, Dr. I. Korobkov, Prof. Dr. S. Gambarotta Department of Chemistry, University of Ottawa 10 Marie Curie, Ottawa, ON K1N 6N5 (Canada) Fax: (+ 1) 613-562-5170 E-mail: sgambaro@uottawa.ca Dr. R. Duchateau Department of Chemistry, Eindhoven University of Technology P.O. Box 513, 5600 MB Eindhoven (The Netherlands) E-mail: r.duchateau@tue.nl Dr. R. Chevalier, Dr. K. Schuhen Lyondellbasell Industries, Polyolefine GmbH Industriepark Hoechst, 65926 Frankfurt (Germany) [**] This research was supported by the Natural Science and Engineer- ing Council of Canada (NSERC), by the Eindhoven University of Technology, and by LyondellBasell. Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/anie.201003465. Angewandte Chemie 9411 Angew. Chem. 2010, 122, 9411 –9414  2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim