Direct Generation of Oxygen via Electrocatalytic Reduction of Carbon Dioxide in an Ionic Liquid

The feasibility of long duration crewed space exploration missions will depend on the supply of consumables to keep the crew alive, including oxygen, water, and food. Air revitalization is a function of the environmental control and life support system (ECLSS) currently employed on the International Space Station (ISS) that recovers O2 from metabolically produced CO2 that is collected from the ISS cabin atmosphere. The current state-of-the-art for O2 recovery onboard the ISS is only capable of recovering up to 54% of the O2 available in respired CO2 and it relies on the resupplied H2O to the ISS for H2 to continue the CO2 reduction reaction. An alternative process for O2 generation is solvated electrochemical CO2 reduction, which is enabled in spacecraft environments by use of an ionic liquid (IL) solvent, electrolyte, and catalysis promoter. An electrochemical CO2 reduction (ECRS) could improve O2 recovery from CO2 up to 70%, or it could function to recover 50% O2 from CO2 without a net water loss, which can be advantageous in missions where in-situ resource utilization (ISRU) of Martian atmospheric CO2 is desired. The goal of this project has been to investigate ILs as solvents, electrolytes, and catalysts with electrochemical cell materials and design parameters to assess the feasibility of generating O2 in a spacecraft environment by electrochemical reduction of CO2 as a standalone process or in parallel with the current state of the art CO2 reduction and O2 generation technology for spacecraft ECLSS. The non-volatile nature of ILs enables their use in space exploration applications, which in turn enables liquid system solutions to CO2 capture and conversion in habitats and in low pressure atmospheres (e.g. Mars). However, IL-systems have not been extensively considered in technology research and development for space applications because their stability and usefulness for CO2 capture and conversion had not been collectively realized until within the last decade. The overall goal was approached from both experimental and high-level system concept perspectives in order to develop a complete picture for how this technology could be implemented, as well as the benefits and challenges that could be encountered through the course of its development. Along the lines of high-level system concept considerations, the objectives were to: 1) Analyze the performance requirements for an electrochemical CO2 reduction system (ECRS) for various crewed space exploration mission scenarios and provide a first order estimate of the system sizing, power requirements, and the theoretical O2 recovery efficiency of the different architectures 2) Evaluate the challenges and opportunities of implementing an ECRS in various missions with relevant upstream and downstream processes and systems, with considerations for microgravity and low-gravity environments And pertaining mostly to the experimental work: 3) Identify, select, and synthesize or procure ILs that may be conducive to supporting CO2 electrolysis, and measure their relevant thermophysical properties in aqueous solutions 4) Parametrically characterize the influence of various ILs in aqueous solutions on the performance of solvated CO2 electrolysis 5) Design and develop a prototype electrochemical reactor capable of operating in conditions relevant to supporting air revitalization needs of a crew in a space habitat In the first two objectives, various system architectures utilizing an ECRS were identified for various possible crewed missions to the Moon, Mars, or a deep space target. The two variables that changed how the ECRS may be best designed and utilized were the availability of CO2 at the destination and the level of gravity present. Reactant, product selectivity, current, requirements for an ECRS in three of the application architectures are shown in Table 1. Table 1. Reactant, selectivity, and current requirements for an ECRS Application Reactant req. by ECRS (mol/CM-d) Selectivity for each product (%) O2 generated by ECRS (mol/CM-d) Total Faradaic current (A/CM) Faradaic current for each product (A/CM) CO2 H2O CO H2 CO H2 ECRS only 23.64 27.36 46.4 53.6 25.50 113.91 52.85 61.05 ISS+ 14.51 - 100 - 7.26 32.41 32.41 - Mars ISRU 51 - 100 - 25.50 113.91 113.91 - The application “ECRS only” describes a case where water and CO2 electrolysis take place in the same reactor at the same time (co-electrolysis) to produce CO and H2, which can be used in a downstream carbon formation reactor. This configuration reduces overall system complexity by combining functions in the same system. The application “ISS+” assumes that the ISS oxygen generator assembly (OGA) and Sabatier reactor, or Carbon dioxide Reduction Assembly (CRA), work in tandem with an ECRS, where the ECRS only reduces CO2 unused by CRA in a water-neutral reaction. And lastly, “Mars ISRU” is used in a scenario where CO2 is removed from the Martian atmosphere and only an ECRS using a water-neutral CO2 reduction reaction is used to provide O2 for a crewed habitat. Such a system could also be described as fitting into so-called “O2 only” Mars ISRU schemes. All relevant units in Table 1 are normalized per crewmember (CM), and so can be scaled to crew size. The power requirements for each of the selected applications are shown in Table 2, using the assumptions of a mission with four crewmembers and a total per-cell voltage of 3 V. Table 2. Power requirements for selected application configurations of an ECRS Application Power with Ecell = 3V (W/CM) Total Power, 4 CM (W) ECRS only 341.7 1,367 ISS+ 97.2 389 Mars ISRU 341.7 1,367 For reference, the practical state-of-the-art appears to come from (Masel et al. 2016): c.a. 50 mA/cm2 current density (current normalized to electrode surface area) with 90% selectivity for CO2 at a 3V operating potential, with a run-time of 4500 h. With these metrics, it is expected that an ECRS sized for “ECRS only” or “Mars ISRU” applications with 4 crewmembers would be approximately the size and mass of the ISS OGA including supporting subsystems (248 kg after suggested improvements from (Takada et al. 2018), ¾ ISS rack), which itself is sized for 11 crewmembers. To reduce this size and mass, improvements in current density while maintaining high selectivity for the desired product are required. For the last three objectives, 5 custom ILs were synthesized and 2 commercial-off-the-shelf (COTS) ILs were procured, making 7 ILs tested in total. Six of the ILs had the trifluoroacetate [TFA] anion with variations in the cation of each IL – each cation was an N-heterocycle with a butyl group and a methyl group: 1-butyl-1-methylpyrrolidinium [bmpyrr], 1-butyl-3-methylpyridinium [bmpy], 1-butyl-2-methylpyrazolium [bmpz], 1-butyl-3-methylimidazolium [bmim], 1-butyl-3-methyl-1,2,3-triazolium [bm-3-tri], 1-butyl-4-methyl-1,2,4-triazolium [bm-4-tri]. The seventh IL, 1-ethyl-3-methylimidazolium trifluoromethanesulfonate [emim][OTf], has been commonly used in literature for electrochemical CO2 reduction and has had some of its thermophysical properties already measured vs. water content, therefore acting as a reference IL for experimental methods and performance of electrochemical systems with the other ILs. The density, viscosity, and ionic conductivity of binary mixtures of each IL and water were measured over a wide range of water concentration (0 to 99% mol% water). Excess molar volume and excess viscosity were calculated and fitted with a Redlich-Kister 4th order polynomial, and conductivity data was fit with the Casteel-Amis fitting equation, allowing property prediction for given water concentrations at the experimental temperature (c.a. 295 K). For the six non-reference ILs, these properties had not been reported in literature previously, and the new data may help in understanding the relationship of IL structures and their properties for designing new ILs. As for the performance of the ILs for electrochemical reduction of solvated CO2, each IL was tested in a small volume, split-compartment, glass electrochemical cell with a 30 mol% solution of IL in water in the cathode compartment. The target CO2 reduction product was CO, as CO production would come from an overall reaction according to 2CO2  2CO + O2, a water-neutral reaction where O2 is the only other net product (produced at the anode). The cathode was therefore a simple silver foil connected to a silver wire, which has good selectivity for CO production from electrochemical CO2 reduction in aqueous solutions. Cyclic voltammetry was conducted for each solution after purging the cell with argon, and then again after purging the cell with pure CO2, allowing for a comparison between reduction currents between a “blank,” inert atmosphere, and an “active,” CO2 atmosphere. In a following experiment, each sample was subjected to bulk electrolysis for 30 minutes with pure CO2 bubbling into solution and the gaseous products being analyzed downstream by non-dispersive infrared spectroscopy and mass spectroscopy. While [bmpy][TFA], [bm-3-tri][TFA], and [bm-4-tri][TFA] either did not show catalytic activity for CO2 reduction or did not have reductive stability in potential regions where CO2 reduction would be expected, [bmpyrr][TFA], [bmim][TFA], and [emim][OTf] all showed selectivity for CO production and modest current densities (as anticipated), whereas [bmpz][TFA] only showed minor activity for CO production (albeit, unanticipated). This enforces the notion that the cation of the IL is major contributor to any catalytic or product-promoting effect in solvated electrochemical CO2 reduction, and provides further information about the types of cations that are viable and those that are non-viable for such a process. Lastly in terms of experimental efforts, a prototype, continuous flow, electrochemical reactor was designed and developed in-house to study how various design aspects could affect the overall performance of the reactor, as well as to investigate a unique design that could enable functional performance of the fluid system in a microgravity environment. The design of the vacuum-assisted product removal (VAPR) electrochemical CO2 reduction reactor is described in detail in a recent conference publication (Holquist et al. 2018). In short, the reactor was designed to electrochemically reduce solvated CO2 at a porous gas diffusion cathode coated in silver nanoparticles, where one side of the cathode was in contact with the CO2 rich electrolyte, and the other side was in contact with a lower pressure vacuum source that would remove gaseous products through the cathode as they were generated. The intent of this design was to limit bubble formation and entrainment in the electrolyte (which would require downstream gas-liquid separation in a real system), as well as produce a pure stream of gaseous product (e.g. 99% CO) on the vacuum side of the reactor (to be handled by downstream collection, analysis equipment, or reactors). An entire test bed was designed and built around the reactor in order to study its performance, including liquid recirculation, command and data handling, and gas handling and analysis subsystems. While testing with the entire system was limited, the prototype reactor was verified to function as designed – carbon monoxide was produced at the cathode, removed by the vacuum source, and measured at the gas analysis equipment (which also implies oxygen was produced at the anode as the only likely counter reaction). However, either the efficiency of the vacuum removal was low, the effectiveness of the CO production was low (instead yielding H2), or both of the low efficacy results occurred at the same time. In summary, the concept of the reactor design was experimentally proven to work, but improvements are needed to make it a viable technology. Other reactor styles and relevant experimental work are summarized in the dissertation that show that solvated electrochemical CO2 reduction systems are improving to the point where they should be considered for future research and development related to space applications.