"Electrochemistry Meets Big Data: How novel high-throughput electrochemical characterization methods can provide new insights into reversible protonic ceramic electrolyzer/fuel cells"
The fundamental relationships between performance metrics, impedance spectra, and electrochemical processes in electrolysis devices and other electrochemical systems are highly complex and often difficult to disentangle. This lack of knowledge hinders identification and evaluation of physicochemical processes, limits understanding of how materials and device architecture influence performance, and ultimately inhibits principled design choices. There is a need for practical, robust procedures to gain physical insight into electrochemical device performance. This requires effective design of experiments to sufficiently characterize varying conditions within reasonable time constraints. In addition, new methods for coherent analysis of large electrochemical datasets are necessary to extract meaning from vast amounts of raw data.
Here, we introduce a novel approach to accelerate electrochemical characterization with standard instrumentation by utilizing rapid measurements in both the time and frequency domains. This “hybrid electrochemical impedance spectroscopy” (HEIS) method provides excellent resolution across a broad range of timescales while decreasing measurement time by more than an order of magnitude compared to conventional electrochemical impedance spectroscopy. The technique can be applied to quickly construct detailed electrochemical maps of energy conversion devices across multiple measurement condition dimensions, revealing physicochemical relationships that are hidden in sparse conventional datasets.
We first apply this approach to the combinatorial investigation of electrocatalytically active positive electrode materials in the Ba(Co,Fe,Zr,Y)O3-d (BCFZY) compositional family, aiming to understand how changes in the oxide composition correlate with the changes in electronic properties and electrochemical performance. In total, more than 2,500 impedance spectra are analyzed, representing 432 distinct BCFZY compositions synthesized by pulsed laser deposition and measured at three temperatures under two different gas atmospheres, enabling a new scientific insight into the trends governing electrochemical performance. Our combinatorial experiments demonstrate that Co-rich compositions achieve the lowest overall polarization resistance under both dry air and humid N2, while high Fe content may impede the performance at decreased temperatures. Combinatorial results are supported by isotope-labeled SIMS trace diffusion studies as well as by symmetric and full cell protonic-ceramic fuel cell and electrolysis testing of selected compositions of particular interest. Hierarchical Bayesian analysis indicates that the performance-limiting process depends on the chemical composition, measurement temperature, and atmospheric humidity. This work provides a composition map of condition-dependent electronic properties of materials in the BCFZY perovskite family for their application as air electrodes for protonic ceramic fuel cell and electrolysis applications.
We then apply our approach to comprehensively characterize complete reversible protonic ceramic electrolyzer/fuel cell (PCEC) devices under different temperature, atmosphere, and bias conditions. Here, more than 20,000 distinct HEIS experiments are performed altogether, resulting in a rich dataset that can be assembled to form relaxation hypersurfaces - multi-dimensional analogs of the one-dimensional distribution of relaxation times (DRT) - which are then processed with new analysis techniques to reveal the underlying processes that govern device performance. This approach describes the electrochemical behavior of PCECs with an unprecedented level of detail made possible by accelerated measurement and scalable data processing strategies.
These studies represent just a few of the possibilities enabled by high-throughput electrochemical impedance spectroscopy. The underlying principles and techniques demonstrated here can be adapted to develop various new ways to examine a wide variety of energy-conversion devices, reducing the time required to understand electrochemical materials and device architectures, and thereby enabling a faster design cycle.
