Ni- and Ni-Sn Anode Porous Layers for SOFCs Operating with Carbonaceous Fuels

Abstract

Nickel metal layers with various morphologies have been investigated as electrical contact, electrocatalyst, and catalyst layer for high temperature hydrogen electro-oxidation reactions and heterogeneous reactions, such as steam methane reforming and water gas shift, in high temperature fuel cells supplied by carbonaceous fuels of various nature. From a heterogeneous catalysis perspective Ni metal alloy, with various dopant concentrations, have been studied to increase metal Nickel tolerance to carbon deposition, which is thermodynamically favoured in SOFC fed with a steam/methane molar ratio <2.0. Metallic nickel electric and thermal conductivity are typically of the order of 107 S/m and 102 W/m.K, respectively, while its coefficient of thermal expansion measures 12x106 m/m.C , making it a good and relatively cheap candidate for electric contact layers and current collectors. The optimization of its morphology is fundamental to improve the nickel layer mass transport properties in conjunction with its catalytic and electrocatalytic performance. Others have proposed and optimized Ni-based porous layers: Smorygo et al. investigated nickel metallic foam supports to promote steam methane reforming and found good corrosive resistance and stability1. Fu et al. studied the effects of modifying nickel foam with copper; they found that copper modification significantly reduced the amount of carbon deposition found in the foam after operation2. Enhanced stability of the foam was reported alongside the resistance to coking. Kan and Lee fabricated a tin-doped Nickel/YSZ anode catalyst which was proven to drastically improve cell stability at a comparable performance3. The cell performance and deposition rates were reported as highly dependent on the tin concentration.Our goal is to compare the electrochemical performance, the catalytic activity towards steam methane reforming reaction yield, and the resistance towards the Boudouard reaction for solid carbon formation of a Ni-based SOFC: i) without Ni porous layer; ii) with a 200 μm thick Ni porous mesh; iii) with a 200 μm thick NiSn porous mesh (porous nickel and nickel-tin mesh ‘Celmet’ are supplied by Sumitomo Electric Toyama Co., Ltd.). We investigate the electrocatalytic performance of the two Nickel-based porous layers on a Ni-YSZ/Ni-GDC|SSZ|LSC-YSZ button cell with 20mm diameter (active area 1.23 cm2). The porous meshes are characterized by a pore size of 230-250 μm (as depicted in Figure 1(v)). The cell assembly is depicted in Figure 1 (iv), where we show how the Ni porous mesh works both as a catalyst layer and a mass transport layer between the gas bulk phase and the triple phase boundary.The cells are operated between 700°C and 850°C with inlet feed compositions comprising CH4, H2, H2O, and N2 with varying S/C ratios from 2.5 to 0 (i.e., dry methane feed). The performance of the three configurations is expressed in terms of electrochemical impedance spectroscopy (EIS) and chronoamperometry measurements within the temperature and composition experimental matrix. X-ray diffraction (XRD) and fluorescence (XRF) are used to analyze the cells and meshes at beginning of life and post-mortem to identify any carbon phases forming on either surface. Scanning Electron Microscopy Energy Dispersion Spectroscopy (SEM-EDS) is used to identify carbon composition and map both the cell and mesh surfaces.The electrochemical performance between case i) and case ii) is reported in Figure 1(i), where polarization curves are summarized for a cell operating with a 1:1 mixture of H2/H2O at 750°C. The increased polarization of case ii) is associated to the additional series resistance of the Ni mesh on the anode layer. Preliminary operation with mixtures of CH4/H2O with S/C as low as 1.5 display performances in-line with expectations, while long-term degradation effects due to carbon deposition remain to be studied. XRD measurements report the formation of solid carbon on a Ni mesh operating with S/C=1.5 (see Figure 1(ii)) for a cell working at 750°C for less than 50 hours. From our preliminary results, we show that the addition of a porous nickel mesh results in methane reformation prior to reaching the cell anode. The deposition is characterized by XRD and SEM and reported along with the polarization curves and electrochemical impedance spectroscopy which define the performance of the cells. O. Smorygo et al., Int J Hydrogen Energy, 34 (2009).X. Z. Fu et al., Int J Hydrogen Energy, 35 (2010).H. Kan and H. Lee, Appl Catal B, 97 (2010). Figure 1