Mathias Schulze

Dry layer preparation and characterisation of polymer electrolyte fuel cell components

Abstract The main problem for future fuel cell commercialisation is the cost of membrane–electrode assemblies (MEAs) satisfying both power density and lifetime requirements. At DLR, low-cost MEA production techniques are being developed. These new MEAs are characterised and investigated with physical and electrochemical methods in order to study the power loss processes, the lifetime, the reaction mechanisms and in support of MEA development. The possibilities for the characterisation methods used will be demonstrated by various examples. At DLR, a new production technique based on the adaptation of a rolling process is developed for fuel cell electrode and MEA preparation. After mixing the dry powder electrode material in a mill, it is blown onto the membrane (or backing) resulting in a uniformly distributed layer. This reactive layer is fixed and thoroughly connected to the membrane by passing them through a calender. In order to produce the second electrode, the same steps are repeated. This procedure is very simple and, as a dry process, avoids the use of any solvents and drying steps. We have achieved a thickness of the reactive layer as low as 5 μm, reducing the amount of catalyst needed and, thus, the costs. Electrochemical investigations have shown a performance comparable to that of commercial electrodes. The degradation of MEA for polymer membrane fuel cell (PEFC) components during the cells lifetime, yields a change in the electrochemical behaviour. The characterisation of PEFC MEA-components after electrochemical operation has given information about the degradation of electrodes and membranes and about the change in the platinum distribution on the anode, whilst on the cathode, the platinum content is unchanged.

Change of electrochemical impedance spectra during CO poisoning of the Pt and Pt–Ru anodes in a membrane fuel cell (PEFC)

Abstract The influence of carbon monoxide poisoning on the platinum and platinum–ruthenium anode in a polymer electrolyte fuel cell was investigated using electrochemical impedance spectroscopy (EIS). EIS is a very useful method for the characterisation of fuel cells. Therefore, impedance measurements of the cell under constant load were performed at periodic time intervals. Due to the poisoning effect of the carbon monoxide, the system changes its state during the experiment. The reconstruction of quasi-causal spectra was made possible using enhanced numerical procedures, especially the time course interpolation and the Z-HIT refinement. The reconstructed impedance spectra show a strong time dependence and exhibit pseudo-inductive contributions at the low-frequency part of the spectra which increase during the experiment. The analysis of the spectra suggests that the pseudo-inductive behaviour can be attributed to a surface relaxation process of the anode. Furthermore, the influence of the carbon monoxide on the electrochemical behaviour of the contaminated fuel cell may be interpreted by means of a Faraday impedance in addition to a potential-dependent hindrance of the charge transfer.

Photoelectron study of electrochemically oxidized nickel and water adsorption on defined NiO surface layers

Abstract Electrochemically oxidized nickel and the water adsorption on NiO prepared in UHV were studied by XPS, TPD and UPS. Potassium ions are intercalated in the electrochemically oxidized nickel surface. No molecular water inserted in the electrochemically formed NiOOH layer was observed after transfer to UHV. Two reduction steps can be distinguished during the reduction of the electrochemically oxidized nickel surface in H2 atmosphere. The interaction of water with NiO depends on the surface structure. The TPD signal of the monolayer desorption of H2O on NiO(100) demonstrates two different adsorption states. The water does not dissociate on the smooth NiO(100) surface as well as on the defected NiO(100) surface. In contrast on the polar NiO(111) surface approximately 0.5 ML (monolayer) water dissociate to OH and H after starting water adsorption.

Investigation of the degradation of different nickel anode types for alkaline fuel cells (AFCs)

Low temperature fuel cells have the opportunity of becoming important for mobile energy systems as, in contrast to other low temperature fuel cells, the alkaline type requires neither noble metal catalysts nor an expensive polymer electrolyte. In alkaline fuel cells (AFC) nickel is used as anode catalyst in gas diffusion electrodes. The metal catalyst was mixed with polytetraflourethylene (PTFE) as organic binder in a knife mile and rolled onto a metal web in a calandar to prepare the electrode. After an activation process with hydrogen evolution at 5 mA/cm2 for 18 h the electrodes were stressed at constant loading in a half cell equipment. During the fuel- cell operation the electrochemical performance decreased. dDue to changes of the polymer (PTFE) and of the metal particles in the electrode, which is described in detail in another paper. In this study three types of electrodes were investigated. The first type of electrode is composed of pure Raney-nickel and PTFE powder, the nickel particles in the second electrode type werewas selected according to particle size and in the third electrode copper powder was added to the nickel powder not selected by size. The size selected nickel particles shows a better electrochemical performance related to the non-selected catalyst, but due to the electrochemically induced disintegration of the nickel particles the electrochemical performance decreases stronger. The copper powder in the third electrode is added to improve the electronic conductivity of the nickel catalyst, but the copper is not stable under the electrochemical conditions in fuel cell operation. With all three anode types long-term experiments have been performed. The electrodes haves been characterized after the electrochemical stressing to investigate the degradation processes.

XPS analysis of PTFE decomposition due to ionizing radiation

X-ray induced photoelectron spectroscopy (XPS) in combination with depth profiling has been used to investigate the structure and the degradation mechanism of PTFE bonded gas diffusion electrodes (GDE). The XP-spectra of these electrodes show distinctly separated binding states of the C1s electrons at Eb=292 eV and Eb=286 eV. These binding states are related to the carbon in the (CF2)n configuration (C1sCF2) and the graphite (C1sgraphite), respectively. The C1sCF2 signal decreases are induced by both X-ray exposure and ion etching. Simultaneously a decrease of the F1s signal has been approved. The intensity ratio of F1s to C1sCF2 has increased during the experiment. These results indicate a decomposition of PTFE which creates CF fractions, leading to an excess intensity in the energetic range between the C1s binding states of the PTFE and the graphite. Although both the F1s and the C1s spectra are strongly modified by ionizing radiation, samples are comparable, when exposition doses are equal.

Investigation of Membrane Pinhole Effects in Polymer Electrolyte Fuel Cells by Locally Resolved Current Density

In order to increase the reliability of fuel cells, an online diagnostic method for detection of operation malfunctions, as well as the early detection of failures in the fuel cells, is necessary. For this purpose, locally resolved current density measurements are an important tool, but the interpretation of the data related to the detection of malfunctions or failures is not straightforward. Here, segmented cell technology is applied to investigate the current density distributions in the anode and cathode electrodes to ascertain their equivalence due to the strong perpendicular coupling of currents. Current density distributions are further used to determine the signature of pinhole formation in the membrane. Different behavior is observed for membrane leakage under open circuit and under applied load conditions. Whereas the cell at open circuit is characterized by a positive current in the vicinity of the pinhole and small negative currents in the remaining area, an applied load leads to large negative currents at the pinhole. The characteristic behavior can be explained by high crossover rates of hydrogen from the anode to the cathode. The nongeneric signature is used to detect the deterioration of a membrane electrode assembly after a test stand malfunction. A sudden pressure drop associated with vaporation of water and the fast cooling of the cell is assumed to trigger the failure of the membrane. After 48 h, fissures in several positions of the membrane near the edges of the cell holder are observed. Through the evolution of leakages in the fuel cell, a malfunction can be detected at an early stage and thereby catastrophic failure of the whole stack may be avoided or anticipated.

Activation of nickel-anodes for alkaline fuel cells

In alkaline fuel cells (AFC) electrodes containing porous nickel and PTFE can be used as anodes. A low-cost production technique yields passive, polymer covered electrodes, which have to be activated before being used in electrochemical devices. During this activation process the surface structure as well as the chemical composition and the oxidation state of the used materials are changed by a given electrochemical current. The electrodes are investigated in different states during this activation process with X-ray photoelectron spectroscopy, scanning electron microscopy with energy dispersive X-ray spectroscopy and nitrogen adsorption.

Interaction of Water with Clean and Oxygen Precovered Nickel Surfaces

The adsorption of water on a Ni(111) single crystal surface, clean as well as precovered with oxygen, has been investigated with thermal desorption spectroscopy (TDS) and measurements of the adsorption-desorption equilibrium combined with XPS (X-ray photoelectron spectroscopy). The measurements have been carried out with water pressures up to 10−5 mbar on surfaces, which have been either clean or precovered with oxygen. On the clean Ni(111) surface the first adsorbate layer with a maximum coverage of 0.42 ML (monolayers) has a desorption energy of 52 kJ/mol and a preexponential factor of desorption of 1016s−1. A second water layer adsorbs with the desorption energy of the ice multilayer but with first order kinetics. On Ni(111) precovered with chemisorbed oxygen an additional state of molecular, more strongly bound water is found, but no dissociation. For higher oxygen precoverages where NiO islands are formed on the surface, also the water dissociation product OH is found adsorbed. On a sample covered with a closed NiO layer, adsorbed OH and molecular water in an energetically not well-defined state are found. High doses of water on oxygen-precovered Ni(111) induce a slow surface modification leading to water dissociation.

The kinetics of the adsorption and desorption of the system CO/Pt(111) derived from high resolution TREELS

We have studied the kinetics of adsorption and desorption of CO on Pt(111) in the temperature range between 430 and 220 K using time-resolved electron energy loss spectroscopy (TREELS). A quantitative analysis of the nonequilibrium data yield the main energies describing the adsorption potentials of the two adsorption sites.

New dry preparation technique for membrane electrode assemblies for PEM fuel cells

A major problem for future fuel cell commercialisation is the cost of membrane electrode assemblies (MEAs) satisfying both power density and lifetime requirements. New, low-cost MEA production techniques are being developed at the DLR (the German national aerospace research centre). These new MEAs have been characterised and investigated using physical and electrochemical methods to support the MEA development. This technique, based on the adaptation of a rolling process, has been developed for fuel cell electrode and MEA preparation. The procedure is very simple, and as a dry process it avoids the use of any solvents and drying steps. We have achieved a thickness of the reactive layer as low as 5 urn, reducing the amount of noble metal catalyst needed to less than 0.05 mg/cm 2 , and thus reducing the costs.