Preparation and Characterization of Components for Intermediate Temperature Fuel Cells And Electrolyzers

The intermediate temperature region for fuel cells (200-400°C) is of interest as it may combine advantages from low and high temperature technologies. Increasing the temperature above what is used in polymer electrolyte membrane (PEM) fuel cells enhances the catalyst kinetics, and therefore it might become possible to use non-noble metal catalysts.

On the other hand, the temperature is low enough for a wide range of materials to be used as construction materials. In this work a set-up was built and fuel cell hardware was made for demonstration of fuel cells for the intermediate temperature range. For the electrodes, carbon cloth and carbon paper were tested as gas diffusion layers with different catalytic compositions, and of the two, carbon paper with a platinum loading of 7 mg cm−2 had the better performance.

However, carbon is unstable at the conditions in the fuel cell cathode and other materials must be sought for. It was attempted to use stainless steel (316L), this however corroded and therefore a protective tantalum coating was applied. The tantalum coatings were found to be corrosion resistant and furthermore provided extremely low interfacial contact resistances of only 1.3 mΩ cm2 .

From a literature review it was found that the most promising results for this temperature range have been performed using cesium dihydrogen phosphate (CsH2PO4) electrolytes. CsH2PO4 undergoes a phase transition at around 230°C, with a rise in conductivity from 8.5 x 10−6 at 223°C to 1.8 x 10−2 S cm−1 at 233°C this is called superprotonic.

This electrolyte as well as other electrolytes for this temperature range, however, suffers from poor mechanical properties, and stable fuel cell performance had only been achieved by use of thick electrolytes. Furthermore to maintain high conductivity of the electrolyte, a high level of humidification was necessary.

Composites with CsH2PO4 were made to improve the properties of the electrolyte material. Composites in formation with mechanically strong materials including ZrO2, TiO2 and NdPO4·0.5H2O improved the densification of the electrolyte, which further resulted in improved stability of the fuel cell. Open circuit voltages (OCVs) using such fuel cells were found to be high, above 0.9 V, and stable up to 250°C.

Composite formation with ZrO2 furthermore resulted in increased conductivity at higher temperatures probably due to the physical stabilization of the high conducting phase. At 250°C the cell was stable for more than 60 hours with a partial pressure of water of only 0.12 atm, and it was operational up to 275°C, where the fuel cell using pure CsH2PO4 no longer performed.

When CsH2PO4 was used in composite with NdPO4·0.5H2O there were indications of a new phase formed, CsH5(PO4)2, which has been reported to have high conductivity from 150°C. The mechanism behind an increase in conductivity for the CsH2PO4/NdPO4·0.5H2O ofvi several orders of magnitude was not fully clarified.

Using an 29CsH2PO4/71NdPO4·H2O electrolyte enabled fuel cell performance measurements up to 285°C, where the highest performance was recorded. At this temperature current and power densities were found to be 117 mA cm−2 and 27.7 mW cm−2 , respectively. Composite formation with melamine cyanurate resulted in increased conductivity in the entire temperature interval measured i.e. from 120°C to 260°C.

A conductivity as high as 0.18 S cm−1 was measured for a 90CsH2PO4/10melamine cyanurate composite at 250°C. Good mechanical properties were furthermore observed for the composites. Within the research project a screening was made in order to search for new electrolytes. From this screening niobium and bismuth phosphates were found to have high conductivities (>10−2 S cm−1 ) with reasonable stability, and it was therefore attempted to fabricate electrochemical cells from these.

The pure phosphates were however suffering from poor mechanical stability and therefore polybenzimidazole (PBI) was added. By adding high amounts of PBI stable OCVs were achieved, these remained stable for around 10 and 70 hours for niobium and bismuth phosphates, respectively. At high temperatures, however, the OCVs were found to drop, at 200°C the OCVs were below 0.9 V.

Tungsten carbide was evaluated as a non-noble catalyst for the hydrogen evolution and oxidation reactions. Tungsten carbides were prepared in different ways in order to achieve higher surface areas compared to the very low surface area of the commercial carbide which was too low to be quantified. By preparing the carbide from WO3 (WC-mWO3) which had been prepared by use of a mesoporous silica template by carburization with methane at 900°C for 3 hours, a surface area of 6 m2 g −1 was measured.

By introducing an extra synthesis step by first converting the WO3 into W2N which was then converted into WC (WC-mW2N) a higher surface area of 18 m2 g −1 was measured. The use of methane versus ethane as carburizing agents were investigated, by carburizing commercial WO3 with both agents under the same conditions.

From carburization with methane no surface area could be quantified, while the carburization with ethane resulted in a carbide (WC-ethane) with a surface area of 12 m2 g −1 . An additional tungsten carbide (WC-05-VN) with a BET area of 31 m2 g −1 was used for comparison. Hydrogen evolution activities for the carbides were measured in phosphoric acid at 185°C and -100 mV.

It was found that apart from WC-mW2N, the activities were increasing with surface area, this deviation may be due to an amorphous carbon surface layer. Activities were found as 1.5, 2.07, 10.7 and 18.73 A g−1 for WC-mWO3, WC-mW2N, WC-ethane and WC-05-VN, respectively. The carbides were furthermore investigated as fuel cell anode catalysts.

The best performances were achieved at the highest temperature measured i.e. 270°C where power densities of 2.7, 3.1, 7.4 and 8.2 mW cm−2 for WC-mW2N, WC-mWO3, WC-05-VN and WC-ethane, respectively, using CsH2PO4 electrolytes and WC loadings of 10 mg cm−2 .