Novel anodes for solid oxide fuel cells

Solid oxide cells are high-temperature electrochemical cells that offer a highly efficient conversion of chemically stored energy into electrical energy and reverse. The current state of the art solid oxide cells use a Ni/YSZ cermet fuel electrode. However, these electrodes experience significant degradation during unintended redox cycles, are prone to carbon deposition, and even small amounts of often seen impurities in the fuel, like sulfur, will inhibit the fuel electrode functionality. Thus, alternatives to Ni-cermet electrodes and especially nickel-free fuel electrodes, are still receiving considerable interest. One strategy to overcome some of the challenges with the Ni/YSZ electrodes is to decouple the functionalities of the fuel electrode by separating the electronically conductive phase from the electrocatalytic phase. The concept of decoupling these two phases from each other is utilized in these studies.

Before utilizing the decoupling of functionalities, a general introduction to the technology and its challenges is given. A screening of the electrocatalytic activity of six different transition metals, namely cobalt, nickel, iron, copper, molybdenum, and wolfram, was performed. These transition metals were infiltrated as nanoparticles into a porous Nb-doped SrTiO3 backbone material. The obtained results showed that copper, molybdenum, and wolfram have insufficient electrocatalytic activity towards the fuel reactions. The cobalt, nickel, and iron, on the other hand, showed decent and comparable electrocatalytic activity in both 4% H2O/H2 and 50% H2O/H2. The iron nanoparticles were found to be the better electrocatalyst at low pO2 values; however, at higher pO2 values, the iron became the inferior catalyst.

Investigating the temperature dependence of the polarization resistance in terms of activation energy and pre-exponential factor showed significant differences between the metal catalysts and a dependence on the gas atmosphere. The results were analyzed by developing an empirically model for the adsorption process, based on the harmonic oscillator. The developed model allows for correlating changes in activation energy and pre-exponential factor with kinetic parameters of the electrode reaction. The model indicates that at higher pO2 values, iron nanoparticles experience a surface blockade due to a strong bonding of the reactants to the catalytic sites. While for the cobalt and nickel electrocatalyst, an increase in steam content, according to the model, leads to an increase in the turnover frequency, which is in good agreement with earlier reports on nickel in literature.

Combining different metals can give rise to synergistic effects achieved from the respective favorable properties of the respective base-metal and secondary-metal. Combinations of the three transition metals, cobalt, nickel, and iron, were therefore investigated. The bimetallic infiltrations were infiltrated alongside gadolinium doped ceria (CGO), in order to introduce ionic conductivity into the structural matrix. Not only combinations of two metals, co-infiltrated with CGO, were investigated, but also pure metals, co-infiltrated CGO, in order to recognize the influence of combining two metals. It was found that the electrocatalytic activity of the bi-metallic infiltrations was similar to the mono-metals with polarization resistances of 0.35 Ω cm2, 0.28 Ω cm2, and 1.32 Ω cm2 for the Co-Fe-CGO, Ni-Co-CGO, and Ni-Fe-CGO infiltrated cells, respectively, at 750oC in 50% H2O/H2.

Raman spectroscopy is a powerful tool for investigating surface chemistry and, when combined with electrochemical impedance spectroscopy under in situ / operando conditions, the technique can report the real-time material composition of the electrode during the impedance measurements. This experimental technique was utilized for investigating the coking resistance of the most promising electrocatalytic materials and material compositions. The technique was applied to five different infiltrations, namely nickel, cobalt, Ni-CGO, Co-CGO, and Ni-Co-CGO infiltrated electrodes. Data showed that all nickel-containing electrodes are prone to carbon deposition when exposed to pure methane; however, by introducing 25% CO2 in the fuel stream, these electrodes become coking resistant down to 750oC. On the contrary, no carbon deposition was found to take place on cobalt, and Co-CGO infiltrated electrodes. Under operando conditions, the electrochemical data suggested that nickel-containing electrodes experience an inductive loop. This inductive loop is a consequence of CO2 being formed during the electrochemical reactions, CO2, which runs the dry reforming process. Moreover, it was found that the cobalt-containing electrodes experienced an additional process below 0.1 Hz, a process that could potentially be the oxidation of solid carbon into CO. These synchronous in situ / operando Raman-EIS studies proved a powerful tool for investigation of the carbon tolerance of these novel fuel electrode materials.