(645b) Orders of Magnitude Decrease In Dye Diffusion In Nanoconfined Polymer Films: Fluorescence Nonradiative Energy Transfer Technique

Upon nanoscale confinement, properties of supported polymer films, such as the glass transition temperature (Tg), often deviate from those of the bulk material. In confined, supported polymer films, a reduction in Tg has been associated with enhanced mobility near the air-polymer surface while an increase in Tg has been connected to attractive polymer-substrate interactions that are stronger than free-surface effects. Polymer properties other than Tg can also be impacted by confinement, such as mobility associated with diffusion. Compared to the many previous Tg-nanoconfinement studies, relatively few studies on diffusion in confined polymer systems have been reported.

Here we demonstrate the dramatic impact of confinement on the out-of-plane diffusion of dyes in supported polymer films. By employing a fluorescence nonradiative energy transfer/break-through time method, we have determined the diffusion coefficients of the dyes Disperse Red 1 (DR1), 9,10-bis(phenylethynyl)anthracene (BPEA) and decacyclene  in polystyrene (PS) as a function of supported film thickness. When film thickness is reduced from bulk to 60 nm, up to a three-order-of-magnitude decrease in the diffusion coefficient of the dyes is observed in PS films at Tg,bulk + 3 K. There is a greater reduction in the diffusion coefficient of the smaller dyes upon confinement, with a factor of 10 decrease in the diffusion coefficient of decacyclene to a factor of 1000 decrease in the diffusion coefficient of DR1. The onset thicknesses for deviation from bulk diffusion coefficients are also greater for the smaller dyes, ranging from 150 nm for decacyclene to 450 nm for DR1.  These film thicknesses are all above the onset thickness for Tg reduction in PS. Furthermore, while PS exhibits a reduction in Tg as film thickness decreases, associated with enhanced mobility, dye diffusion coefficients decrease upon confinement. Therefore, we conclude that there is no direct correlation between Tg-confinement effects and dye diffusion-confinement effects.

These results can be explained by the fact that Tg reflects the average, i.e., longer time side, of the relaxation time distribution of cooperative segmental mobility while probe diffusion reflects the shorter time side of the relaxation time distribution. We hypothesize that confinement suppresses the shorter time side of the relaxation distribution, resulting in the decrease in dye diffusion coefficients, and also the longer time side of the relaxation distribution, leading to the decrease in Tg. This agrees with our findings that the smallest dye exhibits the largest onset thickness for diffusion coefficient reduction and also the largest diffusion coefficient reduction upon confinement.