The recent discovery of the ability of salicylaldiminato Ni(II) complexes to promote ethylene polymerization creates a potential for a new class of olefin polymerization catalysts. The major advantage of this type of catalyst is that they produce a neutral active center and thereby avoid the ion-pairing problems encountered with the homogeneous single-site catalysts in current use. The present DFT study investigates the polymerization mechanism of these neutral complexes as well as the electronic and steric effects of various substituents on the catalyst backbone. The addition of electron-withdrawing or -releasing substituents on the 5 position of the salicylaldiminato ring was found to result in small changes in the energies of the reactions in the polymerization mechanism. This is most likely due to the substituents' remoteness from the active center. Changing the electronic nature of the donor atoms resulted in larger shifts in energy. Finally, bulky substituents such as 2,6diisopropylphenyl and 9-anthracenyl groups were found to have the largest effect on the reaction barriers and enthalpies in a direction that should substantially increase catalyst activity.

Calculations have been carried out on the reaction between the olefin polymerization precatalyst (1,2-Me 2 Cp) 2 ZrMe 2 (P) and a number of Lewis acids (A) to form the methidebridged (contact) ion pair (1,2-Me 2 Cp) 2 ZrMe(µ-Me)A (I; P + A f I + ∆H ipf ). This is the first step in the activation of P to becoming an olefin polymerization catalyst. The calculated enthalpies of formation (∆H ipf ) for I were as follows (A, ∆H ipf in kcal/mol): 1d), -18.0; B(C 6 H 5 ) 3 (1e), -6.7; B(C 10 F 7 ) 3 (1f), -25.8; (MeAlO) 6 (2a), -15.9; (MeBO) 6 (2b), -22.3; AlMe 3 (2c), -8.1; Al(C 6 F 5 ) 3 (2d), -30.8. The charge separation between the (1,2-Me 2 -Cp) 2 ZrMe + and AMefragments in I was calculated for all A, and it was found that the charge separation as well as -∆H ipf increases through the series 1e, 1c, 1b, and 1a with the number of fluorine atoms. A good activating Lewis acid (A) has the equilibrium shifted strongly from P and A toward I, and this is the case for all A except 1e and 2c. Also considered was the complete dissociation in solution (toluene) of I into the counterions [(1,2-Me 2 -Cp) 2 ZrMe] + and AMewith the dissociating enthalpy ∆H ips as well as the formation from I of the solvent-separated ion pair (S ) toluene) [(1,2-Me 2 Cp) 2 ZrMe] + -S-[AMe]with the reaction enthalpy ∆H ss . The two types of separation processes have both been postulated as the second and final steps in the activation of P. The calculated values are as follows (A, ∆H ips and ∆H ss in kcal/mol):

The enthalpy of activation by B(C6F5)3 and subsequent ion pair formation for the mono(cyclopentadienyl), constrained-geometry, and bis(cyclopentadienyl) titanium and zirconium precatalysts were investigated by DFT methods. Solvation effects were incorporated by single-point calculations using the conductor-like screening model, and where appropriate, a single molecule of the solvent was included to model the short-range solvent−solute interactions. The enthalpy of methide abstraction to form a contact ion pair was exothermic for all systems investigated, and the electron-donating ability of the ligands around the metal center has the most predominant effect on its magnitude. Subsequent studies focused on the reactions of this contact ion pair with the olefin and the solvent (toluene). The insertion of ethylene between the cationic and the anionic moieties was found to be an endothermic process for all six catalyst precursors investigated. The bis(cyclopentadienyl) systems showed the least endothermic ethylene complexation energy of 6.1 and 8.2 kcal/mol for the titanium and zirconium precatalysts, respectively. The insertion of toluene between the contact ion pair was found to be exothermic for both the mono(cyclopentadienyl) and the zirconium constrained-geometry catalysts but endothermic for the titanium constrained-geometry catalyst and both of the bis(cyclopentadienyl) catalysts. The bulky ligands hinder the approach of the toluene in the bis(cyclopentadienyl) systems, making olefin complexation the more competitive route. The strong tendency for the mono(cyclopentadienyl) and constrained-geometry systems to coordinate with toluene may be an obstacle to olefin complexation.

No abstract

Abstract:The photo-NOCAS reaction that combines methanol, sewing as the nucleophile, and the radical cation of 4-methyl-1,3-pentadiene (14+'), substituting on the 1,4-dicyanobenzene radical anion (I-'), yields (E)-I-(4-cyanopheny1)-4-methoxy-4-methyl-2-pentene (15) as the major product. This regioisomer arises from bonding of methanol to C-4, the more heavily alkylsubstituted carbon of the diene, giving the less alkyl-substituted allylic radical. All previous examples of the photo-NOCAS reaction have yielded major adduct(s) having regiochemistry consistent with the anti-Markovnikov rule; the more heavily substituted (more stable?) P-alkoxyalkyl radical was the predominant intermediate. Empirically derived heats of formation and high-level ab initio molecular orbital calculations (MP216-3 1G*//HF/6-3 lG*) provide convincing evidence that of the two alternative allylic radicals, generated upon addition of methanol to 14+', that which has the more alkyl substituted allylic radical moiety is, in fact, not the more stable. Of course, the total structure of the intermediate must be considered; the stabilizing effect of alkyl substitution on the carbon next to the oxygen of the ether moiety cannot be ignored. Ab initio molecular orbital calculations (MP216-3 1 G*//HFl6-3 lG*) are reported for the radical cations of 2-methylpropene (2+'), 2-methyl-2-butene (6+'), 2-methyl-l,3-butadiene (9"). 4-methyl-l,3-pentadiene (14+'), and 2,4-dimethyl-l,3-pentadiene (18"). Calculations were also carried out on possible intermediates (bridged radical cations, distonic radical cations, and P-alkoxyalkyl radicals) involved upon reaction of these radical cations with methanol. Results of these calculations provide a basis for explaininglpredicting the regiochemistry of the photo-NOCAS reaction involving methanol as the nucleophile: the major adduct(s) result(s) from attachment of methanol to that end of the alkene or diene which gives rise to the more stable intermediate radical. The more stable radical is not necessarily the more heavily alkyl substituted.Key words: photoinduced electron transfer, radicals, radical cations, ab initio molecular orbital calculations.RCsumC : La rCaction de photo-NOCAS permet de combiner le mCthanol, qui agit comme nucltophile, et le cation radical du 4-mtthylpenta-l,3-dikne (14+'), qui se substitue sur l'anion radical du 1,4-dicyanobenzkne (I-'), et fournit du (E)-1-(4-cyanophCnyl)-4-mCthoxy-4-m&hylpent-2-2 (15) comme produit principal. Le rCgioisomkre rCsulte d'une liaison du mCthanol au C-4, le carbone du dikne le plus substituC par des groupes alkyles, qui conduit au radical allylique le moins substitut par des groupes alkyles. Tous les exemples de rCactions photo-NOCAS rapportCs antkrieurement ont toujours fourni des produits majoritaires dont la rtgiochimie Ctait en accord avec la rkgle anti-Markovnikov correspondant i un intermkdiaire predominant comportant le radical P-alkoxyalkyle le plus substituC (le plus stable?). Les chaleurs de formation que l'on peut obtenir empiriquement de m&me que des c...