Polymer electrolyte fuel cells (PEFC) convert the chemically bound energy in a fuel, e.g. hydrogen, directly into electricity by an electrochemical process. Examples of future applications are energy conversion such as combined heat and power generation (CHP), zero emission vehicles (ZEV) and consumer electronics. One of the key components in the PEFC is the membrane / electrode assembly (MEA). Both the membrane and the electrodes consist of proton conducting polymers (ionomers). In the membrane, properties such as gas permeability, high proton conductivity and sufficient mechanical and chemical stability are of crucial importance. In the electrodes, the morphology and electrochemical characteristics are strongly affected by the ionomer content. The primary purpose of the present thesis was to develop experimental techniques and to use them to characterise proton conducting polymers and membranes for PEFC applications electrochemically at, or close to, fuel cell operating conditions. The work presented ranges from polymer synthesis to electrochemical characterisation of the MEA performance. The use of a sulfonated dendritic polymer as the acidic component in proton conducting membranes was demonstrated. Proton conducting membranes were prepared by chemical cross-linking or in conjunction with a basic functionalised polymer, PSU-pyridine, to produce acid-base blend membranes. In order to study gas permeability a new in-situ method based on cylindrical microelectrodes was developed. An advantage of this method is that the measurements can be carried out at close to real fuel cell operating conditions, at elevated temperature and a wide range of relative humidities. The durability testing of membranes for use in a polymer electrolyte fuel cell (PEFC) has been studied in situ by a combination of galvanostatic steady-state and electrochemical impedance measurements (EIS). Long-term experiments have been compared to fast ex situ testing in 3 % H2O2 solution. For the direct assessment of membrane degradation, micro-Raman spectroscopy and determination of ion exchange capacity (IEC) have been used. PVDF-based membranes, radiation grafted with styrene and sulfonated, were used as model membranes. The influence of ionomer content on the structure and electrochemical characteristics of Nafion-based PEFC cathodes was also demonstrated. The electrodes were thoroughly investigated using various materials and electrochemical characterisation techniques. Electrodes having medium Nafion contents (35 45 wt %), the cathode becomes limited by diffusion of O2 both in the agglomerates and throughout the cathode. Furthermore, models for the membrane coupled with kinetics for the hydrogen electrode, including water concentration dependence, were developed. The models were experimentally validated using a new reference electrode approach. The membrane, as well as the hydrogen anode and cathode characteristics, was studied experimentally using steady-state measurements, current interrupt and EIS. Data obtained with the experiments were in good agreement with the modelled results.

The influence the composition of the cathode has on its structure and electrochemical performance was investigated for a Nafion content spanning from 10 to 70wt.%. The cathodes were formed on a Nafion membrane by the spray method and using 20wt.% Pt on Vulcan (E-TEK). Materials characterisation (SEM, STEM, gas and mercury porosimetry, electron conductivity) and electrochemical characterisation (steady-state polarisation curve, impedance spectroscopy in O2 and current-pulse measurements in N2) were performed. The impedance spectra were analysed using our dynamic agglomerate model. The results indicate that the agglomerate model is valid until a Nafion content of about 45wt.%. Pt/C and Nafion are homogeneously mixed for any composition and no Nafion film was observed. The cathodes containing 36–43wt.% Nafion display a single or double Tafel slope behaviour ascribed to diffusion limitations in the agglomerates. At larger Nafion content, the agglomerate model can describe the curves only by assuming a diffusion coefficient 3–4 decades smaller than that of gases. At such compositions, the porosity was only 10%. These results were interpreted as a blocking of the pores and a non-percolating pore system for too large Nafion contents.

A new method to study permeation of gases in proton conducting membranes using a cylindrical microelectrode is presented. The focus of this work was to develop an in-situ method to study transport properties of hydrogen and oxygen close to real fuel cell operating conditions. The gas permeability is strongly affected by the change of water content in the membrane and it is therefore of advantage that, by using this method, measurements can be carried out over a wide range of relative humidities. The numerical method makes it possible to separate the diffusion coefficient and the concentration of dissolved gas in the membrane and also allows kinetic limitations to be taken into account. Chronoamperometric measurements on Nafion® 117 were successfully evaluated numerically. Experiments at temperatures of 25 and 60°C with respect to oxygen permeation and at 60°C for hydrogen permeation at relative humidities in the range 30–94% are presented. The reproducibility of data was excellent when measuring with different microelectrodes on the same membrane sample, but differed when measuring on different samples. In general, the permeability increases with increasing temperature and relative humidity.

The loss of chemicals in static fish tests due to biodegradation has been studied with eight model surfactants whose biodegradabilities in the OECD Screening Test range from 0 to 99 % BiAS removal. There was a good correlation between biodegradation in the fish test and in the OECD Screening Test. Only surfactants with a very high BiAS removal of above 93 % were also degraded quickly in the fish test. Surfactants with less than 80 % BiAS removal incurred losses of less than 20 % in the fish test. These results indicate that those substances which are not lost by volatility or adsorption will be maintained to at least 80 % of their nominal concentration during a 96 hour static fish test.