Silicon-Containing Polymers for High-Temperature Applications, 18-9108

Printer Friendly Version

Principal Investigator
Mark E. Van Dyke
Michael A. Miller

Inclusive Dates: 01/01/99 - 07/01/01

Background - Poly (silarylene siloxanes) (PSS) are a class of polymers that offer excellent mechanical, thermal, and oxidative properties. Mechanical strength, high heat resistance, and oxidative stability are particularly desirable for aerospace applications. Early development of PSS was intended for applications during the "space race" of the 1960s. This unique class of materials, employing organic and inorganic constituents in the same polymer backbone, demonstrated excellent properties in early testing. Unfortunately, the synthetic techniques developed for production of the monomeric precursors, known as silphenylenes, were not particularly efficient. Consequently, this class of polymers has not achieved any appreciable commercial success. The fundamental challenge in the production of silphenylenes is the creation of a silicon-phenyl bond in high yield under relatively mild conditions. Traditionally, silicon-carbon (Si-C) coupling reactions have been performed in the phenyl system using aromatic Grignards or organolithium compounds. These types of reactions do not lend themselves to large-scale commercial production, so a suitable technique for the synthesis of silphenylenes must avoid these schemes. Recently, Si-C coupling, and silphenylene production specifically, has been achieved using metallocene catalysts. Metallocenes are organometallic compounds that have enjoyed widespread success in the production of both organic and silicon polymers. It is only logical then, that these catalysts would demonstrate activity toward Si-C coupling.

Approach - The objective of this project was to evaluate several metallocene catalyst systems for activity toward Si-C coupling. Of particular interest is the dehydrogenative Si-C coupling of aromatic organics with functional silanes that can then be converted to PSS precursors. The approach taken was to use computational techniques to select from potential systems a few catalysts that would be evaluated in laboratory experiments. When the computational approach yielded only partial success, catalysts were then selected according to other criteria.

Accomplishments - The data obtained in this study suggest that metallocene catalysts with hafnium, niobium, and platinum metal centers have activity toward Si-C dehydrocoupling. Unfortunately, the catalysts display an apparent greater activity toward hydrosilation, a potentially competing reaction. Of these metal centers, hafnium in the form of bis-(pentamethylcyclopentadienyl) hafnium dichloride showed the most activity. However, niobium may be the most promising metal studied, owning to the fact that methyl substitution of the cyclopentadienyl ligand appears to increase overall activity in this system. A nonmethyl-substituted niobium catalyst yielded 51 and 7.4 percent of hydrosilated and dehydrocoupled product, respectively. Methyl substitution of the niobium catalyst's ligands would be expected to improve upon the yields achieved in this study with a nonmethyl-substituted niobium catalyst. To completely explore this phenomenon, experiments using bis-(pentamethylcyclopentadienyl) niobium dichloride catalyst should be conducted. Further experiments using bis-(pentamethylcyclopentadienyl) hafnium dichloride would also be worthwhile, particularly in light of the discovery by other researchers that 3,3-dimethylbut-1-ene as a co-catalyst improves yields of Si-C dehydrocoupled products. A system that produces a mixture of compounds such as the one under investigation here could be useful without separation of reaction products. If the overall yield of di-substituted products is high, aromatization techniques such as heating over platinum catalyst can be employed to convert all disilated products back to aromatics. Meta-, and para-derivatives can then be separated if necessary. Overall yields of Si-C dehydrocoupled products may also be increased through the use of a cocatalyst such as 3,3-dimethylbut-1-ene.

2001 IR&D Home SwRI Home