SWITCH project at the SOFC-XVII

The project partner EPFL (École polytechnique fédérale de Lausanne), on July 2021, participated in the digital meeting of the 17th International Symposium on Solid Oxide Fuel Cells (SOFC-XVII).

The event, sponsored by the High-temperature Energy, Materials and processes division of the Electrochemical Society, Inc. and the SOFC Society of Japan, saw the participation of the SOC community from academia and industry. ECS has indeed a large network of over 8,000 scientists and engineers.

The two presentations given by EPFL, within the scope of the SWITCH project, were

1- Local Characterization of a Solid Oxide Cell Operated in Fuel Cell and Electrolysis Mode Using Lock-in Thermograph

Abstract:

Operando characterization of SOCs generally relies on current–voltage measurements that provide the spatially averaged electrochemical response of the SOC. However, these methods do not fully capture the spatial distribution of the operating conditions (e.g., gas composition, current density). Local information on the operating conditions can be gained by using operando optical characterization methods. In this work, the spatial distribution of the electro-thermal response of an SOC with a 15 cm2 active area operated in fuel cell and electrolysis mode was investigated using lock-in thermography. Lock-in thermography consists of stimulating the solid oxide cell with a sinusoidal current perturbation and analyzing the local thermal response using Fourier transforms. A camera with a silicon-based sensor was used to record the transient temperature field on the oxygen electrode (i.e., the cathode in fuel cell operation) via a specifically designed optical access. The field of view of each pixel was approximately 0.03 mm x 0.03 mm. The Fourier transform generates a compact data set that can be conveniently represented by a phase and amplitude or a real and imaginary image. Using a real–imaginary representation provided a better thermal response contrast than an amplitude–phase representation. The local thermal response was found to be dependent on the direct current bias and the frequency of the current perturbation. The thermal response of the SOC was laterally and longitudinally inhomogeneous, thereby demonstrating the capability of active thermography to characterize the local behavior of an SOC in operando. Lock-in thermography is thus a promising method for defining regions of interest for post-mortem analysis using electron microscopy and helping to link highly localized post-test analysis with spatially averaged operando characterization methods.

More information at this link.

2 – Modeling Nickel Microstructural Evolution in Ni-YSZ Electrodes Using a Mathematical Morphology Approach

Abstract:

Microstructural evolution of Nickel in Ni-YSZ fuel electrodes is one of the main limiting degradation mechanisms in solid oxide cells [1]. The redistribution of Nickel is ascribed to its high mobility and low wettability with YSZ, related to interfacial surface tensions. Practically, the degradation manifests as the agglomeration of Ni, potentially accompanied by its migration, mainly reported in electrolysis mode.

Extensive efforts have been made to understand Ni agglomeration and relocation. In this frame, advanced imaging techniques have been employed to characterize the microstructure of electrodes after aging in a wide range of conditions. In order to support these observations and gain fundamental insights, a number of computer simulations methods such as: multistate kinetic Potts-Monte Carlo models, cellular automata, phase field, and topological boundary dynamics have been successfully employed in modelling grain growth and recrystallization phenomena. Nevertheless, questions still remain on the exact underlying mechanisms of Ni migration.

In the shadow of this lack of clear understanding, a morphological modelling approach aiming at mimicking Ni depletion based on the most fundamental definition of chemical potential and surface curvature minimization would be very helpful to guide researchers into gaining clearer insights of this phenomenon. Moreover, the morphological simulations are very fast compared to the physically based models. Therefore, they can offer the possibility to emulate large database of numerical microstructures in order to study the impact of Ni evolution on the microstructural properties [2].

The current work falls within this framework and aims at modelling the two main Ni degradation phenomena, namely coarsening and migration. For this purpose, a novel model based on morphological operations from mathematical morphology [3] has been developed and adapted to the Ni-YSZ specificities. The considered driving force for Ni evolution is the minimization of its surface curvature, and implicitly its chemical potential, which strongly depends on the local operating conditions. After calibration, the model showed a capability to predict, starting from an initial 3-D microstructure as input data, Ni relocation that was in agreement with that observed in a large dataset of 3-D reconstructions from pristine and aged cells. This helps understanding the associated microstructural evolutions which are linked to the cell Area Specific Resistance (ASR) through an adapted in-house model. This coupling of microstructural and electrochemical models allows giving clearer insights and practical recommendations for the design of fuel electrodes with improved stability and electro-catalytic activity.