(559a) Polymer Dynamics In Single Wall Carbon Nanotube / Polystyrene Nanocomposites

Polymer nanocomposites provide access to new regimes of polymer dynamics in which the impenetrable filler particles are comparable to and frequently smaller than the polymer chain conformation. In this study, single wall carbon nanotubes (SWCNTs) / polystyrene (PS) nanocomposites were prepared by a coagulation method. Using this method, the nanotubes are evenly distributed in the polymer matrix and the SWCNTs exist as small bundles with a diameter of 9.6 nm and aspect ratio of ~35. An elastic recoil detection (ERD) method was employed to detect the tracer, deuterated polystyrene (dPS), diffusion into the nanocomposites and the tracer diffusion coefficients were determined. Rheological properties were measured in the PS linear viscoelastic regime.

The tracer diffusion coefficient first decreases and then increases with increasing SWCNT loading. The polymer entanglement molecular weight and the relaxation time remain constant according to our linear viscoelastic measurements. Therefore, this does not explain the change in the tracer diffusion coefficient. The transition from decreasing to increasing tracer diffusion corresponds approximately with the onset to rheological percolation and appears to increase to a higher SWCNT loading with decreasing matrix molecular weight. A trap model is proposed to describe the polymer diffusion in these nanocomposites. SWCNT bundles introduce traps around them into the composites system. Within and out of the trap, the polymer diffusion rates are the same, but the diffusion slows down when the tracer moves away from the nanotubes. When the SWCNT loading is below the percolation threshold, the traps are isolated and the diffusion coefficient decreases. With increasing SWCNT loading, the SWCNT bundles percolate and the traps overlap, which allows the tracer polymer to move along the nanotube network and the apparent tracer diffusion coefficient increases. With further increasing SWCNT content, the volume fraction of the trap in the composites becomes larger and larger, the tracer diffusion coefficient recovers back to the diffusion coefficient in the pure polymer. Simulated results from the trap model compare favorably with our experimental results.