| dc.description.abstract | Proton Exchange Membrane (PEM) Fuel Cells are one of the most promising futures for green energy, as they provide a sustainable alternative to combustion engines in vehicles. The bipolar plate (BPP) is a crucial component in the PEM fuel cell, providing mechanical support to the stack as well as ensuring the even flow of reactants into the cell and waste products out of the cell. Additionally, it is an important conductive medium for the transfer of electrons through the stack, and any losses occurring in the BPP will reduce the overall performance of the fuel cell.
Metallic materials for bipolar plates are nowadays favoured for their high conductivity and mechanical strength. However, given the harsh conditions inside a PEMFC, they will form a protective oxide on the surface, which reduces the cell performance over time due to an increase in interfacial contact resistance (ICR). Dissolution of the metal bipolar plate could also result in leaching of harmful metal ions which cause irreversible damage to the membrane. Therefore, metallic BPPs must be coated with a conductive and stable material to increase the lifetime and performance of the fuel cell.
This thesis explores a new concept of soldering the gas diffusion layer (GDL) to the BPP using metallic tin (Sn). A thin layer of Sn is formed on a stainless steel BPP using electrodeposition, and then hot pressed with the GDL at a temperature close to the melting point of Sn. This leads to excellent conduction pathways through the BPP to the GDL, drastically reducing contact resistance and improving fuel cell performance. Additionally, the Sn coating will oxidise under the conditions present inside a fuel cell, leading to a protective SnO2 oxide layer which should prevent further corrosion whilst maintaining conductivity through the soldered GDL fibres. This soldering process produces a BPP/GDL system with an excellent performance and longevity inside a fuel cell, using a simple forming method and low-cost materials.
BPPs produced in this way were optimised to give the lowest contact resistance and best corrosion resistance, as judged from electrochemical testing in a simulated PEMFC environment. The optimised Sn/GDL BPP was found to perform much better than stainless steel (SS316) alone, with an ICR of 6.5 mΩ cm2 and 13.2 mΩ cm2 obtained before and after corrosion testing, respectively. The increase in ICR after electrochemical testing was attributed to the dissolution and precipitation of the Sn to form a SnO2 layer, which was not completely protective.
Additional modifications to the Sn/GDL system can be made to improve the stability of the protective oxide layer, and the interface between the Sn and the GDL. Both the addition of bismuth and indium as alloying elements were investigated, with additions of up to 4 % bismuth improving performance. Additions of more than 4 %, and excessive heat treatment of the samples during hot-pressing led to the phase separation of the alloy which had a detrimental effect on performance. The presence of a carbon nanolayer at the Sn/GDL interface also enhanced the conductivity of the system and specifically multiwalled carbon nanotubes maintained a high conductivity even after exposure to a simulated PEMFC environment.
Three of the different Sn/GDL concept bipolar plates were tested inside a working fuel cell to compare to standard stainless steel and gold coated stainless steel bipolar plates. All three Sn/GDL, SnBi/GDL, and Sn/C/GDL BPPs performed better than stainless steel alone, with a higher cell voltage and lower ICR at all points during the 200 hours (600 drive cycles) of testing. The Sn/GDL performed the best of the three overall, with a lower degradation rate and small increase in ICR after testing. Although the SnBi and Sn/C BPPs performed well over the first 500 cycles, the harsh conditions experienced during the shut down and start-up procedures led to enhanced degradation during the final 100 cycles. It was predicted that the performance of the Sn/GDL BPP would meet and even surpass the performance of the Au coated BPP if cycling were to continue.
Finally, the suitability of the Sn/GDL system on an industrial scale was analysed. Sn/GDL as a BPP material was found to be a much cheaper alternative to Au coated BPPs, with an estimated cost of 7 $ / kW compared to over 100 for gold. This is still more expensive than the industrially produced Bpps, which cost around 5 / kW, but with small improvements in performance and large scale production, the Sn/GDL BPP could be a cheaper alternative. Additionally, a unique high-throughput joining method to form fuel cell stacks by using the Sn as a solder material is proposed. This would provide a highly efficient manufacturing method and inevitably lead to further cost reductions.
Therefore, the novel Sn/GDL concept for BPPs is both high performance and low cost, and represents a promising alternative to traditional coated bipolar plates for use in PEM Fuel Cells. | en_US |
