Viscoelastic phase diagram of fluorinated and grafted polymer films and proton-exchange membranes for fuel cell applications

Abstract

The influence of temperature and moisture activity on the viscoelastic behavior of fluorinated membranes for fuel cell applications was investigated. Uncrosslinked and crosslinked ethylene tetrafluoroethylene (ETFE)-based proton-conducting membranes were prepared by radiation grafting and subsequent sulfonation and their behavior was compared with ETFE base film and commercial Nafion® NR212 membrane. Uniaxial tensile tests and stress relaxation tests at controlled temperature and relative humidity (RH) were carried out at 30 and 50 °C for 10% < RH < 90%. Grafted films were stiffer and exhibited stronger strain hardening when compared with ETFE. Similarly, both uncrosslinked and crosslinked membranes were stiffer and stronger than Nafion®. Yield stress was found to decrease and moisture sensitivity to increase on sulfonation. The viscoelastic relaxation of the grafted films was found to obey a power-law behavior with exponent equal to -0.04 ± 0.01, a factor of almost 2 lower than ETFE, weakly influenced by moisture and temperature. Moreover, the grafted films presented a higher hygrothermal stability when compared with their membranes counterparts. In the case of membranes, a power-law behavior at RH < 60% was also observed. However, a markedly different behavior was evident at RH > 60%, with an almost single relaxation time exponential. An exponential decrease of relaxation time with RH from 60 s to 10 s was obtained at RH ≥ 70% and 30°C. The general behavior of grafted films observed at 30°C was also obtained at 50°C. However, an anomalous result was noticed for the membranes, with a higher modulus at 50°C when compared with 30°C. This behavior was explained by solvation of the sulfonic acid groups by water absorption creating hydrogen bonding within the clusters. A viscoelastic phase diagram was elaborated to map critical conditions (temperature and RH) for transitions in time-dependent behavior, from power-law scaling to exponential scaling. © 2013 Wiley Periodicals, Inc.