A Comparison of Molten Sn and Bi for Solid Oxide Fuel Cell Anodes

Molten Sn and Bi were examined at 973 and 1073 K for use as anodes in solid oxide fuel cells with yttriastabilized zirconia (YSZ) electrolytes. Cells were operated under "battery" conditions, with dry He flow in the anode compartment, to characterize the electrochemical oxidation of the metals at the YSZ interface. For both metals, the open-circuit voltages (OCVs) were close to that expected based on their oxidation thermodynamics, ~0.93 V for Sn and ~0.48 V for Bi. With Sn, the cell performance degraded rapidly after the transfer of approximately 0.5-1.5 C/cm2 of charge due to the formation of a SnO2 layer at the YSZ interface. At 973 K, the anode impedance at OCV for freshly reduced Sn was approximately 3 Ω cm2 but this increased to well over 250 Ω cm2 after the transfer of 1.6 C/cm2 of charge. Following the transfer of 8.2 C/cm2 at 1073 K, the formation of a 10 μm thick SnO2 layer was confirmed by scanning electron microscopy. With Bi, the OCV anode impedance at 973 K was less than 0.25 Ω cm2 and remained constant until essentially all of the Bi had been oxidized to Bi2O3. Some implications of these results for direct carbon fuel cells are discussed. Disciplines Biochemical and Biomolecular Engineering | Chemical Engineering | Engineering Comments Suggested Citation: Jayakumar, A., S. Lee, A. Hornés, J.M. Vohs and R.J. Gorte. "A Comparison of Molten Sn and Bi for Solid Oxide Fuel Cell Anodes." Journal of the Electrochemical Society. Vol. 157(3). pp. B365-B369. © 2010 The Electrochemical Society, Inc. All rights reserved. Except as provided under U.S. copyright law, this work may not be reproduced, resold, distributed, or modified without the express permission of The Electrochemical Society (ECS). The archival version of this work was published in Journal of the Electrochemical Society. http://dx.doi.org/10.1149/1.3282443. This journal article is available at ScholarlyCommons: http://repository.upenn.edu/cbe_papers/130 A Comparison of Molten Sn and Bi for Solid Oxide Fuel Cell Anodes A. Jayakumar, S. Lee, A. Hornés, J. M. Vohs,* and R. J. Gorte* Department of Chemical and Biomolecular Engineering, University of Pennsylvania, Philadelphia, Pennsylvania 19104, USA Institute of Catalysis and Petrochemistry, Consejo Superior de Investigaciones Científicas, 28049 Madrid, Spain Molten Sn and Bi were examined at 973 and 1073 K for use as anodes in solid oxide fuel cells with yttria-stabilized zirconia YSZ electrolytes. Cells were operated under “battery” conditions, with dry He flow in the anode compartment, to characterize the electrochemical oxidation of the metals at the YSZ interface. For both metals, the open-circuit voltages OCVs were close to that expected based on their oxidation thermodynamics, 0.93 V for Sn and 0.48 V for Bi. With Sn, the cell performance degraded rapidly after the transfer of approximately 0.5–1.5 C/cm2 of charge due to the formation of a SnO2 layer at the YSZ interface. At 973 K, the anode impedance at OCV for freshly reduced Sn was approximately 3 cm2 but this increased to well over 250 cm2 after the transfer of 1.6 C/cm2 of charge. Following the transfer of 8.2 C/cm2 at 1073 K, the formation of a 10 m thick SnO2 layer was confirmed by scanning electron microscopy. With Bi, the OCV anode impedance at 973 K was less than 0.25 cm2 and remained constant until essentially all of the Bi had been oxidized to Bi2O3. Some implications of these results for direct carbon fuel cells are discussed. © 2010 The Electrochemical Society. DOI: 10.1149/1.3282443 All rights reserved. Manuscript submitted September 21, 2009; revised manuscript received November 30, 2009. Published January 19, 2010. It is theoretically possible to convert solid carbonaceous fuels, including biomass or coal, directly into electricity using fuel cells based on electrolytes that transfer oxygen ions. A major challenge in making these direct carbon fuel cells DCFCs practical is the requirement of fabricating low impedance anodes that allow facile transfer of oxygen from the electrolyte to the surface of the solid fuel. Although this transfer can be accomplished in the gas phase using a CO–CO2 redox couple, 3 the use of electrodes based on liquid metals and molten carbonates would offer improved fuel flexibility. The majority of work in DCFC has used a mixture of moltencarbonate salts e.g., Li2CO3 + K2CO3 + Na2CO3 as the anode, using either a molten-carbonate or a yttria-stabilized zirconia YSZ electrolyte. Although the mixed carbonate salts appear to be very good oxidizers and impressive performance levels have been achieved with this type of anode, molten carbonates are not electronically conductive so that a metallic current collector must be incorporated into the anode structure. In addition to the problems associated with dissolution of metals like Ni into the salt solution, the anode half-cell reaction, Reaction 1, requires transfer of electrons so that only the fuel, which is in contact with the current collector, can be oxidized C + 2O − = CO2 + 4e − 1 The situation is completely analogous to what happens in normal solid oxide fuel cell SOFC electrodes, where reaction can only occur at three-phase boundary sites, that line where the gas phase, the electronic conductor, and the ionic conductor all come together. One method for increasing the region where Reaction 1 can occur involves using a conductive form of carbon as the fuel and maintaining a high concentration of that carbon within the carbonate solution. This obviously limits what fuels can be used because the fuel itself is part of the anode. Another approach for transferring oxygen from the electrolyte to the solid fuel involves the use of anodes composed of liquid metals, such as Sn or Bi. In this case, the metal reacts at the electrolyte interface via Reaction 2, and the metal oxide MO is in turn reduced by the carbonaceous fuel in a separate step M + nO − = MOn + 2ne − 2 For cells using Sn anodes, it has been reported that this type of fuel cell can be operated in a “battery” mode by simply allowing the metallic Sn to be consumed and then be regenerated later. However, while the concept has been demonstrated with liquid Sn anodes, there is very little fundamental information available to show what limits the performance of these electrodes and how one might improve them. In the present paper, we set out to investigate Reaction 2 in a fuel cell with a YSZ electrolyte, using Sn or Bi as the anode. With Sn, we show that a critical issue that limits performance at temperatures below 1073 K is the formation of a SnO2 film at the electrolyte interface due to the very low solubility of oxygen in molten Sn. The performance of the Sn-based electrodes appears to be limited by the low ionic conductance of this SnO2 layer, at least for temperatures below 1073 K. With molten Bi, the electrochemical reaction is facile, probably because the oxide is a good ionic conductor; however, the critical issue with Bi is its lower open-circuit potential.