Journal of Membrane Science, Vol.302, No.1-2, 70-77, 2007
Upper bound relationship for proton exchange membranes: Empirical relationship and relevance of phase separated blends
The investigation of polyelectrolyte membranes for proton conductivity in fuel cell applications has significantly increased in the past decade. The literature now contains a myriad of experimental investigations directed at achieving high proton conductivity in polyelectrolyte membranes primarily based on sulfortated polymers. It has been observed in our studies that the conductivity versus water content yields a linear log-log relationship for specific polymer variants. This suggests an upper bound relationship may exist expressed by a linear log-log plot of the proton conductivity versus water content of the membrane. Literature data plotted according to this protocol demonstrates an empirical upper bound exists. Water sorption is thus the scaling variable and is related to membrane performance as the optimized systems will exhibit a maximum in conductivity achieved with a minimum in water sorption. This is a trade-off relationship as polyelectrolytes will generally exhibit increased conductivity with increasing water sorption as the variables to achieve the increased conductivity (i.e. ion exchange capacity (IEC)) invariably leads to increased water sorption. Another aspect of the investigation reported in this paper involves the phase behavior of polyelectrolyte membranes as the hydrophobic regions may be considered phase separated from the hydrophilic regions as often noted in the case of Nafion (R). This is more apparent in polymer blends or block copolymers comprised of hydrophobic and hydrophilic regions. A number of papers have empirically noted that phase separated morphologies can deliver more interesting proton conductivities compared to the unblended constituent values. The adaptation of simple continuum mechanics and transport models to conductivity models allows the prediction of proton conductivity of phase separated systems. An analysis of these models confirms the empirically observed optimization of conductivity in phase separated systems compared to the homogeneous systems. (c) 2007 Elsevier B.V. All rights reserved.