Use this URL to cite or link to this record in EThOS:
Title: Carbon-air fuel cells with molten tin anodes
Author: Colet Lagrille, Melanie Alejandra
Awarding Body: Imperial College London
Current Institution: Imperial College London
Date of Award: 2013
Availability of Full Text:
Access from EThOS:
Access from Institution:
Carbon-air fuel cells are a prospective technology for efficient conversion of the chemical energy of solid carbonaceous fuels directly to electrical energy. Although CO2 is a product, undiluted by the nitrogen and excess oxygen in typical flue gases, such fuel cells can be considered more environmentally-benign than conventional electrical energy production if carbonaceous wastes are used as fuel. In effect, disposal of this carbonaceous wastes would result in CO2 emissions e.g. from landfill sites, while oxidation of them into a fuel cell would result in energy and neat CO2 which is far more amenable to subsequent treatment. In addition, depending on the operating potential difference of the fuel cell, CO2 emission rates per kWh output can be significantly less than from conventional power stations. Considering that Ni / yttrium-stabilised zirconia (YSZ) anodes in conventional (H2 / O2) solid oxide fuel cells (SOFCs) are irreversibly affected by deposition of carbon and sulphur poisoning, some molten metal anodes offer better alternatives, with high electrical conductivities and stabilities when operating on carbonaceous fuels. Tin (Sn) fulfils these requirements and is non-toxic, has a low melting point and vapour pressure, forms oxide(s) that can be reduced by CO (and H2), and could be purified of ash, based on its immiscibility with molten metals, and of dissolved metals by electrochemical processes. The thermodynamics, kinetics and materials stability of a solid oxide fuel cell with a 90 μm thick lanthanum strontium manganite (LSM)-YSZ cathode, a 2 mm thick YSZ electrolyte and a 5 mm thick molten tin anode (Sn(l)-SOFC) were studied by electrochemical measurements and X-ray analysis. Platinum wires and mesh were used as current collectors at the cathode, while graphite rods or lanthanum strontium titanate (LST) / lanthanum cerium strontium titanate (LCST) pellets were used at the anode. Sn(l)-SOFC performance was tested in the absence of fuel (battery mode) at 900 °C using flow rates of 60 mL min-1 of air at the surface of the cathode and 30 mL min-1 of helium at the surface of the anode. For the Sn(l)-SOFC operating in fuel cell mode, helium or hydrogen were fed into a 10 mm thick anode at flow rates of 30 and 20 mL min-1, respectively, and activated carbon particles were sited on the surface of the melt. Thermodynamic analysis of the open circuit voltages (OCVs) measured in the temperature range between 600 and 900 °C, determined that the global reaction occurring in the Sn(l)-SOFC when operating without fuel was oxidation of molten tin at the Sn anode | electrolyte interface. This result was confirmed by current density transients of a Sn(l)-SOFC operating at its maximum power density (constant potential difference) and post-mortem analysis of the fuel cell components. Accumulation of tin dioxide at the Sn anode | electrolyte and Sn anode | graphite current collector surfaces was responsible for the degradation of the Sn(l)-SOFC longer term performance. The kinetics of the Sn(l)-SOFC operating in absence of fuel were studied using the polarization curves and impedance spectra measured at 900 °C and a mathematical model of the ohmic and polarization losses. The model inputs included kinetic parameters reported in the literature for the electrolyte and cathode, enabling prediction of the anode activation and concentration overpotentials; the model outputs were a charge transfer coefficient of ca. 0.67, an exchange current density of ca. 353 A m-2 and a limiting current density of ca. 3 273 A m-2. These values indicated that the kinetics of the whole Sn(l)-SOFC were limited by the mass transport processes occurring at the anode, since the exchange and limiting current densities associated to the cathode were one order of magnitude higher. Modelling of the Sn(l)-SOFC with a 200 μm thick electrolyte and negligible ohmic losses at the current collectors resulted in a maximum power density of ca. 1 477 W m-2 at a cell voltage of ca. 0.45 V and a current density of ca. 3 273 A m-2, corresponding to the limiting current density. This maximum power density was approximately 1 000 W m-2 lower than the maximum power density predicted for a conventional SOFC with identical LSM-YSZ cathode and YSZ electrolyte (same dimensions and microstructure), but replacing the molten tin anode by a 100 μm thick Ni-YSZ anode operating on hydrogen. Even though the predicted power density of a Sn(l)-SOFC was lower than that of a conventional SOFC, this result seemed to be promising since the Sn(l)-SOFC kinetics could be enhanced by reducing the molten tin anode thickness and improving its contact with the electrolyte by applying pressure over the melt. When the Sn(l)-SOFC was operated in fuel cell mode, the kinetics of tin dioxide reduction by carbonaceous fuels was found to be slow and did not have significant effects in the kinetics of the fuel cell, even when a stirred molten tin anode was used. By contrast, even though the rate of tin dioxide formation at the Sn anode | electrolyte interface was higher than the rate of reduction of tin dioxide by hydrogen, the presence of this fuel improved the anode kinetics. This condition was observed in the results reported by CellTech Power Inc. using a quiescent molten tin anode and the results obtained during this research project using a stirred molten tin anode, which suggested that hydrogen oxidation occurred in parallel with tin oxidation at the Sn anode | YSZ electrolyte interface and possibly at the SnO2 | Sn interface.
Supervisor: Kelsall, Geoff Sponsor: Comision Nacional de Investigación Científica y Tecnologica (Chile) ; Government of Chile ; University of Chile
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral