Abstract

Ni-based cathodes have been widely accepted as the state-of-the-art in terms of their superior catalytic activity for the CO2 reduction reaction (CO2RR) in solid oxide electrolysis cells (SOEC). However, they are prone to NiO formation and coking at high conversion rates, resulting in performance degradation and subsequent loss of activity (1, 2). For these reasons, numerous efforts have focused on investigating alternative electrocatalysts to replace Ni-based SOEC cathodes. Perovskite oxides, with a general formula ABO3, are a versatile class of materials that have very good catalytic properties that can be tailored by doping a variety of cations on the A- or B-site of the parent lattice.Extensive efforts in our group have been focussed on a very promising class of perovskite electrocatalysts, namely La0.3M0.7Fe0.7Cr0.3O3-δ (M = Sr, Ca) (LMFCr) that are not only highly active for oxygen reduction/evolution, but also for CO2 reduction (3-5). These materials are mixed ionic electronic conductors (MIEC) that also exhibit excellent compatibility with conventional electrolyte materials and hence can be employed directly as electrode layers without the need for composite electrodes that require mixing with ionic conductors, e.g., gadolinia-doped ceria (GDC) etc. Futhermore, the same electrocatalyst can be used as both the fuel and oxygen electrode which enables the cell to run reversibly (reversible solid oxide fuel cells, RSOFC). This is highly advantageous in terms of simplifying cell design and hence lowering cost, while also mitigating compatibility issues with neighboring cell components and generating long term stability (6), thus making the LMFCr electrocatalyst very promising for commercialization purposes.Of the two LMFCr analogues, La0.3Ca0.7Fe0.7Cr0.3O3-δ (LCFCr) exhibits superior catalytic activity for both the CO2RR and OER and therefore it is the focus of this work. Here, the stability of the LCFCr catalyst under high CO2 conversion to CO has been examined, determining whether these conditions lead to any deleterious effects, such as coke formation. We have answered this question by studying the stability and electrochemical performance of LCFCr in various CO2:CO environments. Furthermore, we have also employed Ni-doped LCFCr to decorate the perovskite surface with in situ exsolved metal nanoparticles (NP) to determine if this would further enhance its catalytic activity and stability. The LCFCr electrode ink was prepared by employing a 1:1 ratio of powder to ink vehicle. Symmetrical cells were prepared by tape casting the LCFCr ink over a ca. 0.50 cm2 area on both sides of 2.5 cm diameter SDC-buffered YSZ/SSZ substrates. Au paste was painted on as the current collector followed by mounting and sealing the cells on an alumina tube with a ceramic sealant. The fuel electrode was then exposed to various CO2:CO ratios (100:0, 90:10, 70:30, and 50:50), while air was supplied to the oxygen electrode at flow rates of 50 ml/min.The electrochemical performance of the cells was determined via electrochemical impedance spectroscopy (EIS) (at a 50 mV ac amplitude and over a range of 65000 – 0.01 Hz) and using cyclic voltammetry (CV at 5 mV/s). The short- and medium-term stability of the cells was determined by monitoring the current at various cell voltages of 1.0 – 1.6 V. The cells showed excellent performance for CO2RR in these gas mixtures, while the CO oxidation activity plateaued as the overpotential was increased. The CO2RR and CO oxidation activity was significantly enhanced when Ni-doped LCFCr with in situ exsolved FeNi alloy NPs was employed as the electrode layers, also showing remarkably stable performance over several hours at 1.2 V without any evidence of coking or delamination up to 90% CO2 conversion, as revealed by postmortem SEM studies. The CO2/CO cells were also studied using synchrotron radiation at the Canadian Light Source in order to determine whether any changes occurred to the LMFCr surface or bulk properties and if any carbon could be detected, especially at very high conversion of CO2 to CO.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call