Generating Large Thermally Stable Marangoni-Driven Topography In Polymer Films By Stabilizing The Surface Energy Gradient

Marangoni forces drive a fluid to flow in response to positional differences in surface energy. In thin polymer films, a difference in surface energy between two coincident liquid polymers could offer a useful route to manufacture topographically patterned surfaces via the Marangoni effect. Previously, we have demonstrated a photochemical method using the Marangoni effect for patterning thin polystyrene films. To generalize the approach, a theoretical model that gives the underlying physics of this process was also developed, which further revealed that low viscosities, low diffusivities, and large surface energy gradients favor rapid evolution of large film thickness variations. However, as described by the Stokes Einstein equation or the Rouse model, low viscosity is generally correlated with high diffusivity in a single-component system. Herein, we report a strategy to decouple film viscosity and diffusivity by co-casting a high molecular weight surface energy gradient creating copolymer (low diffusivity) with a low molecular weight majority homopolymer (high diffusivity and low viscosity), which are miscible with each other. Patterned light exposure through a photomask imposes a patterned surface energy gradient between light-exposed and unexposed regions due to photochemical reactions involving only the low diffusivity component. Upon heating the film to the liquid state, the film materials (primarily the low viscosity homopolymer component) flow from the low to high surface energy regions. This strategy either eliminates or greatly slows dissipation of the prepatterned surface energy gradient while maintaining rapid feature formation, resulting in formation of ca. 500 nm high features within only 30 min of thermal annealing. Furthermore, the formed features are stable upon extended thermal annealing for up to one month. It is found that a ratio of Marangoni forces to capillary forces can provide a predictive metric that distinguishes which scenarios produce features that dissipate or persist.