Ultrathin Nafion Film Behavior Under Simulated Hot-Pressing and Operating Conditions of a PEFC

It is now widely accepted that Nafion ionomer in the polymer electrolyte fuel cell (PEFC) catalyst layer exists as a thin film (<10nm in). The impact of thermal annealing on the Nafion^® thin films is of particular interest for PEFCs because the membrane electrode assembly (MEA) comprising the catalyst layers (CLs) and the membrane is subjected to a hot pressing step by heating to a temperature ranging 125-195 ^oC under elevated pressure. In an operating fuel cell, the catalyst layer and ionomer film within experiences prolonged liquid water exposure. The polymer molecules can undergo rearrangement when the chemical/thermal environment is altered. In internal and surface structure change can also impact the wettability and transport behavior of ultrathin ionomer films. In this work, the effects of thermal annealing and liquid water exposure on Nafion^®thin films is examined.

Recently, we have reported the properties self-assembledultrathin Nafion ionomer film including the characteristic thickness scale (i.e., 4-10 nm) pertinent to PEFC CLs [1]. Film thickness-dependent surface wettability and bulk proton conductivity were notable features of those unannealed films. We also discussed about the surface switchability of ultrathin ionomer film from super-hydrophilic to hydrophobic when the films were subjected to the high temperature annealing [2]. One of the interesting finding was the restricted ionomer mobility with decreasing thickness that required high energy for the switching of surface wettability, generalized as the term “confinement effect”.

We have further investigated bulk proton transport of those thermally annealed ultrathin Nafion films by electrochemical impedance spectroscopy (EIS). It was found that proton conductivity diminished more than one order of magnitude regardless of thin film thickness compare to the unannealed counterpart. The reason of that might be the higher extent of morphological rearrangement and reorientation upon annealing as evident in atomic force microscopic (AFM) measurement. Interestingly, liquid water exposure to such those annealed films regenerate the both surface and bulk properties where prolonged water vapor exposure cannot alter such properties. Moreover, the regeneration process is highly thickness dependent.   

As an example, the proton conductivity of the 55 nm film at 60 ^oC and different RH has been plotted in terms of treatment condition in the Figure 1. The truncated terms - Un-ann, Ann-1^st, WT (24 h) and WT (96 h) have been used for the film treatment conditions - Un-annealed, 1^stAnnealed, 24 h Water treated and 96 h Water treated, respectively. It was found that proton conductivity of 24 h water treated film was very similar to the annealed films in the RH range 20 to 96%. In contrast, the conductivity of 96 h water treated film went back to unannealed film after 96 h water treatment at high RH where it did not recover completely at low RH. On the other hand, the regeneration process for the much thinner 10 nm film is much quicker as happened by 24 h water treatment (not in the fig).   

Surface wettability of the water treated films was also investigated by water contact angle measurement as shown in Figure 2. It is interesting that the surface wettability was also regenerated from hydrophobic to hydrophilic. The process is thickness dependent as well where the regeneration for thinner films is much quicker than that of thicker films. This is completely opposite trend to the annealing effect. However, there is a consistency of surface and bulk regeneration where film thickness is the differentiating parameter.

The presentation will discuss the results of the study in more detail.

Acknowledgements

 

Financial Early Researcher Awards (Ontario Ministry of Research and Innovation) and Natural Sciences and Engineering Research Council of Canada (NSERC).

References

 

1. D. K. Paul, K. Karan, J. Giorgi, A. Docoslis, J. Pearce, Macromolecules2013, 46 (9), 3461–3475

2. D. K. Paul, K. Karan, ECS Trans. 2013, 50 (2), 951-959.