Solid Oxide Electrolysis for Space Applications

The OxEon team provided the Solid Oxide Electrolysis Cell (SOEC) stack for the MOXIE (Mars OXygen In-situ resource utilization Experiment) project for the Perseverance Rover. The SOEC stack on Mars successfully operated 16 times meeting all the operational objectives and producing propellant quality (>99.6% purity) oxygen by electrolyzing Mars atmosphere CO 2 and represents the first ISRU (In-Situ Resource Utilization) demonstration on another planet. [1] ISRU presents the opportunity to reduce payload and launch costs by utilizing resources already available on the moon and Mars. Since MOXIE, the SOEC stack using identical set of materials has been successfully scaled by 35x and incorporated into breadboard demonstration systems for propellant production on both the moon and Mars. [2], [3] Additionally, materials development since MOXIE has resulted in improved performance and capabilities of the nickel-based cathode material. With the support of NASA, OxEon is continuing SOEC materials and hardware development to design, build, and operate a propellant production system capable of producing methane and oxygen from H 2 O and CO 2 . The MOXIE SOEC device was a 0.5% scale demonstration that produced between 6-12 g/hour high purity O 2 (>99.6%). The stack utilized a traditional nickel-based electrode which is susceptible to oxidation by the feed gas (in this case CO 2 ) at the inlet conditions without a reducing species such as carbon monoxide being present in the feed. To avoid oxidation of the cathode, a recycle loop was implemented in the MOXIE system. Since completion of MOXIE, materials development has occurred resulting in a redox tolerant cathode shown to completely tolerate partial and full (i.e., complete oxidation of Ni to NiO before re-reduction) redox cycling, with performance recovery occurring in a matter of minutes using CO generated by the electrolysis reaction. [4] This materials advancement eliminates the need for a recycle stream in future systems. Dry-CO 2 electrolysis using an SOEC stack to produce high purity oxygen was demonstrated on Mars, but subsequent ISRU SOEC work has coupled that capability with H 2 O electrolysis for additional functionality. Stack testing has been done to demonstrate steam electrolysis, dry-CO2 electrolysis, and co-electrolysis with a range of H 2 O/ CO 2 feed ratios all with the same device. In the case of co-electrolysis, the synthesis gas produced has a H 2 / CO outlet ratio that tracks the feed composition allowing for a gas composition tailored to the production of a range of chemicals including methane, methanol, lubricants, and liquid fuels. OxEon developed, fabricated, and relevant condition tested a breadboard system containing an SOEC stack 35x the size of the MOXIE stack, integrated with a methanation reactor. The integrated system was installed and tested in the Mars chamber at the Jet Propulsion Laboratory to demonstrate production of methane and high purity oxygen from H 2 O and CO 2 at relevant Mars conditions. The system demonstrated oxygen production as high as 680 g/hour with a purity >99.996% and methane production as high as 170 g/hour. A second breadboard system was developed, fabricated, and relevant condition tested for lunar propellant production. An SOEC stack 35x the size of the MOXIE stack was thermally integrated with a Balance of Plant (BOP) designed to intake near freezing water (to simulate water extracted from a Permanently Shadowed Region (PSR) on the Lunar surface) and produce propellant H 2 and O 2 . The integrated SOEC/ BOP system was relevant condition tested in a cryo-vacuum chamber at the Colorado School of Mines to simulate operation in a Lunar PSR. The system successfully met nominal operating targets of 1.8 kg/day H 2 production and 14.4 kg/day O 2 production with a specific energy as low as 48 kWh elec /kg H2 . Additionally, testing was done to demonstrate the ability to electrochemically pressurize the produced oxygen product up to pressures as high as 3.7 bar as a benefit to downstream processing and storage. Remote operation of SOEC system for Mars and Lunar applications imposes a significant demand for performance stability and reliability. Each of the materials set used in the construction of the stack has been evaluated for potential improvement for robust operation. The redox tolerant fuel electrode previously developed under a NASA SBIR project has continued to be utilized for redox capability (Figure 1, left) but also due to its ability for higher conversion of CO 2 when compared to the heritage (MOXIE) fuel electrode. Optimization of the redox tolerant fuel electrode is ongoing but recent materials development has also focused on enhanced electrolyte, air electrode, and barrier layers for improved long-term cell stability. Electrolyte performance and stability during cell operation is a function of the crystallographic phase(s) present at the beginning (thermal history) and any phase changes during operation. Zirconia electrolyte is expected to have a combination of cubic and tetragonal phases; however, a detrimental monoclinic phase can also form that affects the robustness of the cells during stack sealing and eventual performance. An investigation determined that exposure to certain temperature ranges should be limited to avoid monoclinic formation, and changes made to thermal processing resulted in more durable cells. The air/O 2 side of the cell contains an LSCF-based perovskite electrode along with a ceria barrier layer between the electrode and electrolyte. The barrier layer prevents detrimental interactions between the electrode and electrolyte materials, namely the formation of resistive lanthanum zirconate and strontium zirconate, that can take place during cell processing and operation. The ceria barrier layer also forms a resistive solid solution with zirconia electrolyte if the sintering temperature is too high. Multiple improvements were made to the barrier layer to increase cell performance and long-term stability, including use of a sintering aid to reduce the sintering temperature, printing optimizations to prevent barrier cracks, and post-processing steps to increase layer density. Unlike lanthanum, strontium has a strong tendency to migrate from the electrode to the electrolyte boundary through any gaps/cracks or regions of high porosity in the barrier layer. Therefore, a high performance strontium-free electrode is also under investigation to avoid the strontium migration and degradation problem altogether. The combined approach of improved barrier layer and strontium elimination from the electrode layer will be tested for long-term durability in button cell testing. Developments in air electrode treatments have shown that the initial performance of the electrode can be fully recovered, even after 5,000 hours of operation. Baseline electrode treatment results in full performance recovery and improved stability. A modified treatment applied to an air electrode symmetric button cell (half-cell) has demonstrated full performance recovery again, and with significantly improved stability (Figure 1, right). The 0.4-volt hold used for the test (OCV = 0 with air on both sides) results in fuel cell mode of operation for one air electrode and electrolysis mode of operation for the opposite air electrode. Typically, an initial break-in period of about 300 hours is required before reaching stabilized performance. After three thermal cycles to room temperature, around 800 hours test time to verify cycling stability, the cell was cooled for baseline treatment. The cell was re-heated and tested for an additional 4,800 hours until the previous stable performance of 0.3 A/cm 2