The major goals of this work were to: develop a new methodology for the preparation of site-isolated catalytic sites on a silica surface, prepare the first truly single-site supported metallocene/CGC polymerization catalyst, and develop structure-reactivity relationships for these new systems. To synthesize these novel catalysts, the approach taken was to develop a protocol which allows for the synthesis of an aminosilica material with isolated, uniform amine sites. This patterned aminosilica was then used as a scaffold to support a constrained geometry catalyst. These functionalizations occurred at essentially a quantitative level, in stark contrast to previous literature reports. The patterned catalysts were evaluated in the polymerization of ethylene and compared to densely loaded literature materials. Overall, it was found the patterned materials were 5-10 times more active than traditional immobilized CGC catalysts. The patterned catalysts were also found to be effective catalysts for the copolymerization of norbornenes (including functionalized norbornenes) and ethylene, the first reported use of a tethered CGC for the production of ethylene-norbornene copolymers. The control materials were inactive in these polymerizations, providing further evidence that the patterning protocol allows for the synthesis of unique highly active, isolated catalytic sites. Various structural components of the immobilized CGC developed in this work were tested for their impact on catalyst synthesis and reactivity in ethylene polymerizations. The results showed the patterned materials in general behaved according to the trends seen in homogeneous CGC polymerizations. These results, while congruent with similar homogeneous CGC studies, are in direct conflict with previous work on supported CGCs reported in the literature. This discrepancy is likely the result of the difference between the isolated, possibly single-site patterned catalysts developed in the course of this work and the multi-sited catalysts prepared by traditional supporting protocols. This also further illustrates the difficulty in developing structure-reactivity relationships when ill-defined solid catalysts are used.

Utilization of methylaluminoxane (MAO) activated metallocene and constrained geometry (CGC) olefin polymerization catalysts containing fluorenyl or octamethyloctahydrodibenzofluorenyl (Oct) moieties has yielded three series of syndiotactic copolymers of propylene with higher a-olefins. The melting temperatures of these polymers were analyzed, and found to correspond directly with the mole percent incorporation of comonomer, as well as with the frequency of stereoerrors in the polymers. Further analysis indicated that rmrr stereoerrors, a result of site epimerization, occur in close proximity to the incorporated comonomers. The MAO-activated fluorenyl/Oct-containing metallocene and CGC catalysts were further utilized to produce syndiotactic samples of poly(1-butene) (s-PB) and poly(1- pentene) (s-PPe). The syndiotacticity of the samples was quantified by 13C NMR and the melting temperatures determined by DSC. The samples of s-PB and s-PPe produced by Me2Si( h1-C29H36)( h1-N-tBu)ZrCl2·OEt2 (Oct-CGC) were found to melt at higher temperatures (55.9 and 43.1 °C, respectively) than any previously reported samples. The MAO-activated Oct-CGC was also used to produce polyethylene samples at a variety of polymerization temperatures and pressures. All of the samples were found to contain an unprecedented degree of branching (13-65 total branches per 1000 carbon atoms) for an early transition metal single-site catalyst. The branches were found to be almost exclusively of two or greater than five carbon atoms in length, and the levels of the longer branches could be controlled by varying the polymerization conditions. The number of ethyl branches was roughly 5 per 1000 carbon atoms for all samples. Finally, a binary catalyst system comprising the Oct-CGC and a chromium-based ethylene trimerization catalyst, ((tBuSCH2CH2)2NH)CrCl3, was developed. This MAOactivated catalyst system could be tuned to produce polyethylene samples with 17-49 total branches per 1000 carbon atoms. Between 4 and 16 of these branches were found to arise from incorporation of 1-hexene produced by the chromium oligomerization catalyst. Adjusting the ratios of oligomerization catalyst, polymerization catalyst, and activator was found to allow rational control over the branch content of the polymers. The branching levels could also be varied by altering the time between injection of the oligomerization and polymerization catalysts into the system.