The density of Nafion membranes wetted in water and acetonitrile and the solubility of Nafion in a 50:50 ethanol/water mixture are reported. Commercially available membranes, solution recast membranes, and recast membranes thermally processed for varying lengths of time were examined. Unheated recast membranes were found to be ∼20% less dense in water than both commercial membranes and recast membranes heated at 140 °C for 40, 50, and 60 min. For heating times of ≥40 min, the density of recast membranes in water was invariant and equal to the density of commercial films. Similarly, in acetonitrile, commercial membranes and films heated for 1 h had the same density; unheated recast films were ∼25% less dense. Nafion density in acetonitrile was ∼15% less than that in water. Acid pretreatment reduced Nafion film density by ∼15%. Unheated recast films were 22-100% soluble in a sonicated ethanol/water mixture. Commercial membranes and recast films heated for as little as 10 min were insoluble under the same conditions.

Composites are formed by sorption of Nafion, an ion exchange polymer, onto inert substrates of two different microstructures:  microcylinders (neutron track etched membranes) and microspheres. Flux of a neutral redox probe (hydroquinone) through the composites is measured voltammetrically. Data for the microsphere composites are presented for variation in volume fraction and bead radii. Flux is shown to increase as the ratio of surface area of the substrate to volume of Nafion increases. The enhancement is ascribed to a facile surface diffusion domain at the interface between the Nafion and the inert substrates. Enhancements are larger in the microcylinder composites. Images of the microsphere composites appear fractal. For the microsphere composites, the power law relating flux to a characteristic dimension (microsphere radius) has an exponent of 0.78. A similar analysis for the cylinders has an exponent of 1. The analysis also shows that the fundamental transport rates of surface and bulk diffusion are little different in the two composites. Thus, the difference in the flux characteristics of the two classes of composites is ascribed to the microstructure of the interface as opposed to a change in the basic flux properties. (A linear fit of the microsphere composite data is statistically comparable that of the fractal fit; however, the conductivity exponent of one for the linear fit does not capture the dimensionality of the microsphere/Nafion interface.) The smallest microspheres of radius 0.055 μm form agglomerates of radius 1.25 to 1.5 μm, while their flux characteristics are consistent with an effective radius of 0.24 μm.

No abstract

No abstract

No abstract