Dual Function Materials Enabling Human Space Flight: Carbon Dioxide Capture and Conversion for Life Support on Crewed Missions

Abstract

Carbon dioxide removal is important for keeping astronauts alive in space, where CO₂ can accumulate to harmful or even deadly levels in cabin air if untreated. Additionally, on Earth, CO₂ direct air capture is an important technology for reversing the harmful impacts of rising anthropogenic atmospheric CO₂ concentrations. In both scenarios, captured CO₂ needs to be dealt with, potentially via reaction into a more desirable final product such as renewable hydrocarbons or water. One potential solution is utilizing combined solid sorbents and catalysts in one material, known as dual function material (DFM). In this work, DFMs are used to capture and convert CO₂ from spacecraft cabin air into water as a form of recycling, which is necessary for enabling a longer duration human spaceflight. DFM is studied with CO₂ concentrations relevant to cabin air conditions for astronauts (1500 to 3000 ppm of CO₂) both with and without moisture present. DFM CO₂ capacity increases by nearly a factor of 4 and uptake rates by 10 with more realistic moist inlet air compared to dry cabin air. The wet capacity of DFM is comparable to state-of-the-art sorbents in use on the International Space Station (ISS) now; however, ISS systems must dry cabin air before CO₂ capture since they lose CO₂ capacity with a wet air inlet. DFM shows promise to save significant mass, size, power, and complexity for a CO₂ removal and conversion system, which could help enable longer duration human space missions.

1. Introduction

The inhospitable environment of space makes keeping astronauts alive quite a challenge, from the lack of air/pressure to the extreme temperature changes. Engineering solutions put in place are called life support systems (LSS), and they do everything to keep astronauts alive in an enclosed spacecraft environment like maintaining breathable air, handling liquid and solid waste, and regulating environmental conditions like temperature and pressure.¹ The LSS required for space flight are similar to those in other closed environments like submarines; however, there are some unique system limitations for space flight that must be considered like partial gravity, radiation, and launch loads.²

Maintaining breathable air is a very important part of an LSS since losing atmospheric pressure or oxygen concentration can become fatal in only minutes. In addition to providing oxygen to breathe and maintaining other inert atmospheric gases like nitrogen, exhaled CO₂ must also be separated out of the cabin air, so it does not asphyxiate the astronauts. Lithium hydroxide has historically been used for this separation due to its large capacity for adsorbing CO₂, dating back to the Apollo era.^3,4 One major downside to lithium hydroxide sorbents is that they cannot be regenerated in space. Once the sorbent is saturated with CO₂, it is now waste and cannot be recycled for use again. A single use material is fine for short trips to space; however, it quickly becomes impractical for longer duration human space trips like International Space Station (ISS), lunar base, or Mars transit.⁵ Additionally, oxygen for breathing is generated by splitting water into oxygen and hydrogen gases via electrolysis. Currently, captured CO₂ and hydrogen leftover from splitting water are waste streams on the ISS.⁶

For these longer duration human spaceflights, captured CO₂ must be recycled into useful feedstocks like water. One convenient way to do this is reacting captured CO₂ with the previously wasted hydrogen stream to form water and methane in a Sabatier reaction.⁷ Recycling CO₂ into water can cut the required mass at launch by orders of magnitude for a long duration mission. Critically, this shift toward a closed loop LSS that can recycle CO₂ into water necessitates new CO₂ scrubbing technology in which captured CO₂ can be released from the sorbent for recycling.⁸

The first regenerable CO₂ removal technology to fly as architecture on the ISS was the Carbon Dioxide Removal Assembly (CDRA).⁹ This CO₂ removal system combines a water desiccant bed with a 5A zeolite CO₂ sorbent bed to capture CO₂ from the cabin air. Zeolites have a higher affinity for capturing water compared to for CO₂, so the incoming air must first be dried before going to the zeolite sorbent bed. Once saturated, the zeolite sorbent bed is heated to release the CO₂ and regenerate the sorbent to start the next cycle. There are two sets of sorbent beds that operate out of sync with each other to effectively create a continuous process. There is a compressor and tanks downstream of CDRA to pressurize and store the pure CO₂ product stream before it is fed to a Sabatier reactor for conversion to water and methane.¹⁰ CDRA was able to effectively separate CO₂; however, it had some reliability issues like sorbent dusting that used significant crew time for maintenance.¹¹ CDRA has since been updated to fix the dusting issues, improve heating, and change the sorbent to a 13X zeolite. The updated system is called 4BCO2 and keeps the same basic four-bed capture mechanism as CDRA but with slightly shortened cycle times.^12,13 Other hardware in the CO₂ conversion process has also been refined, including Sabatier in preparing its second generation hardware, and alternate compressors are being explored. Sabatier had some performance issues traced to the volatile organic compounds (VOC) on board the ISS which prematurely poisoned its multimetal catalyst; however, they have been resolved in the second generation redesign with additional gas scrubbing steps.¹⁴⁻¹⁶ Several other technologies are also being explored for use in longer duration missions as alternatives to CDRA and 4BCO2, each of which has its own advantages and disadvantages. A solid amine system has flown on the ISS as a technology demonstration,¹⁷ while full scale liquid amine and thermal deposition CO₂ separation ground systems are also being developed.^18,19

CO₂ capture for LSS is a fundamentally similar problem to direct air capture (DAC) processes which are being researched for use in Earth’s atmosphere. The main differences include slightly higher CO₂ concentrations in the air (2000 vs 450 ppm) and different driving design factors. For example, DAC in Earth’s atmosphere prioritizes the cost of captured CO₂ while DAC for LSS systems prioritizes system mass and power use over cost. Technologies like amine and zeolite sorbents are being studied for both LSS and DAC on Earth.^20,21

Recently, DAC and combined methanation reactions have become of interest as a way to not just capture CO₂ but also convert it into a useful product in a single process.²² This reaction allows for cleaner hydrocarbon fuels that can be used in hard to decarbonize sectors. Combining DAC with methanation into one bed has since been called dual function material (DFM).²³ Combing CO₂ capture with conversion into one packed bed can save significant energy compared to having two separate systems which both need heating.²⁴ This DFM concept has previously been validated with good capture and conversion rates over a metal oxide (commonly Na or K) sorbent and Ru catalyst dispersed in an alumina support.^25,26 The basic operation involves flowing CO₂-laden air across the DFM bed at room temperature, followed by a gas purge to ensure little oxygen is present. The same DFM bed that has just captured the CO₂ is then heated while hydrogen flows over the bed, which converts the previously captured CO₂ to water and methane. DFM has been explored for other reactions and with different modes of operation, like using isothermal beds instead of temperature swing.^24,27 Several capture scenarios have been tested including breakthrough bed capacity, moisture effects, and sorbent cyclability.^28,29 A moderate amount of moisture, like that present in atmospheric air, has been shown to improve DFM CO₂ capacity. Additionally, DFM has shown good repeated cyclic performance in the literature with over 450 h on stream, a key property which is highly necessary for DFM use in LSS.³⁰ More exotic supports like monoliths or 13X can be used for DFM materials as well, both of which have strong histories of performance for both DFM and traditional sorbents.³¹⁻³³ One key reaction distinction between the LSS and Earth systems is the importance of recovering the water product for space use compared to methane. While water is the objective for LSS use, methane is still a useful species as a fuel product or for use in further systems like methane pyrolysis, which can strip the hydrogen away for use as a reactant in the DFM process.³⁴

DFM technology could provide a game changing development for closed loop Environmental Control and Life Support System (ECLSS) that could enable long duration human spaceflight and human settlement on previously inhospitable planets by closing the ECLSS loop. This technology can both capture CO₂ and recycle it back into water in one system compared to the three that it currently takes to do so in ISS demonstrations. New closed loop ECLSS is necessary for longer duration human space flight to recycle water, since it becomes impossible to launch enough water to support human life for long missions. This three-into-one simplification would save significant space, power, and complexity compared to the start of the art for regenerable ECLSS. DFM is similar enough to previously used materials in CDRA and Sabatier that many design lessons (such as sorbent dusting, heating, cooling, VOC scrubbing) can be used to help rapidly advance DFM’s maturity compared to other new, unproven technologies. Additionally, DFM has enhanced performance in the presence of water, which will always be in cabin air, compared to the current state of the art on ISS which fails with even a very low moisture content. Previous work done with DFM for DAC shows significant promise to help close the loop on ECLSS to enable longer duration missions.

This work studies DFM (Na₂O/Ru/Al₂O₃) for use in crewed spacecraft LSS as a combined CO₂ capture and conversion system. Combining CO₂ capture and conversion into one step should allow for an overall smaller combined system that uses less power and decreases complexity compared to the current state-of-the-art LSS in use on the ISS. This work aims to explore the adsorption capacity of DFM across CO₂ concentration ranges relevant to LSS (1500–3000 ppm of CO₂) both without and with water to realistically simulate conditions of spacecraft cabin air. Finally, some initial CO₂ conversion data are shared, and next steps are outlined.

2. Materials and Methods

2.1. Materials

All of the solid chemicals used in this work were obtained from Sigma-Aldrich. Ruthenium on alumina powder (5 wt % Ru, reduced, dry) and pellets (0.5 wt % Ru, 3.2 mm, dry) were used as support and catalyst material as prepared from Sigma-Aldrich (BET SA = 150 m²/g). Sodium carbonate (ACS reagent grade, >99.5%) was used as the precursor for the sorbent on the support. Water used for ion exchange and vapor sorption was deionized via a Milli-Q system. Matheson provided CO₂ gas (UHP, >99.995%) for characterization. Gas mixtures to model cabin air for characterization were 2600 ppm of CO₂ in nitrogen (±2%, certified and provided by Matheson).

2.2. Dual Function Material Synthesis

The DFMs were prepared by using an incipient wetness technique. Alumina supports already came impregnated with a ruthenium precursor, so only sodium needed to be added. A solution of sodium carbonate at the saturation limit for water was prepared with a total volume equal to the pore space of the samples being prepared. This solution was then mixed with the alumina support and allowed to dry in a 110 °C oven overnight (∼16 h) in air. This process was then repeated three total times to achieve the maximum 9% sodium loading. DFMs were then calcined in a 350 °C oven for 4 h to drive off any remaining precursor after all rounds of incipient wetness were finished. Both the pellets and the powder were prepared via the same incipient wetness technique.

2.3. Dual Function Material Capture and Reaction Testing

2.3.1. Carbon Dioxide Adsorption Rates

CO₂ capacity of the DFM from cabin air (2600 ppm of CO₂) was measured in a Surface Measurement Systems (SMS) Dynamic Vapor Sorption (DVS) system. Full details are available on the SMS Web site, but the DVS is essentially a microbalance inside of a gas cell.³⁵ A small amount of sorbent sample (∼40 mg) is loaded into a hanging tray on the microbalance. Gas (CO₂ or cabin air) and volatile liquids (water) can be flowed into the cell via independent MFCs in addition to a vacuum line. Additionally, the sample can be heated to 350 °C.

For the CO₂ capacity measurements, samples were loaded onto the tray and degassed for 6 h at 350 °C (5 °C/min) under a vacuum and then allowed to cool. Next, analysis was done with the model cabin air stream (2600 ppm of CO₂ in N2) flowed into the cell. Various flow rates were tested as well as the addition of water, while mass was measured to evaluate the CO₂ uptake rates and final capacity of the DFM.

2.3.2. Methanation Reaction and Water Production

The same DVS setup was used for the methanation reaction, with CO₂ being captured from the cabin air in the same fashion as described above. However, in these experiments, once the material was saturated with adequate CO₂, the flow was switched from model cabin air to hydrogen, and the sample was heated at 5 °C/min up to 300 °C where it was held. Reaction products were monitored in the gas outlet stream via FTIR (Bruker Omega 5). Once the methanation reaction was complete, gas flow was stopped and the samples were allowed to cool.

2.4. Characterization Techniques

2.4.1. Carbon Dioxide Adsorption

CO₂ adsorption tests were run in a Micromeritics ASAP2020. Prior to analysis, a 6 h degassing step at 350 °C (5 °C/min ramp) under a vacuum was used to prepare the samples. Then, analysis was done with pure CO₂ dosing from 0 to 101.3 kPa while the temperature was held at 25 °C.

2.4.2. Carbon Dioxide and Water Coadsorption

The ASAP2020 does not allow for multigas analysis so coadsorption tests were run in the SMS DVS system described previously (2.3). The DVS system allows for a gas stream and water to be dosed up to 100 sccm via independent MFCs. The water is pulled from the vapor space above a liquid water cell, so it can only be dosed at pressures up to its vapor pressure at the given analysis temperature. For the coadsorption experiments in this work, the sample was degassed for 6 h at 350 °C (5 °C/min ramp) prior to dosing with combinations of water and CO₂ gas at flow rates between 0 and 20 sccm, varying for each component. Pressures from 0 to 101.3 kPa at a temperature of 25 °C were used for this analysis.

An FTIR is set up in line on the outlet of the DVS. The constant FTIR sampling allowed for the composition of the outlet stream to be measured to discern monitor reaction progress. The FTIR pressure and temperature matched the conditions in DVS throughout each experiment. Any IR active species in the product gas stream can be identified and quantified with the large species library provided by Bruker.

2.4.3. X-ray Diffraction (XRD)

XRD was run on powdered samples in a Rigaku Smartlab XRD instrument with a Cu Ka source. Measurements were collected from 10 to 80° with a 0.01° step.

3. Results and Discussion

Ruthenium supported on alumina was chosen as a starting point for this work due to the absolute performance of ruthenium compared to other available metals for DFM, despite its higher cost.²³ In aerospace applications, minimizing launch weight is very important, so higher performing materials were chosen rather than the most cost-effective. Commercially prepared Ru/alumina catalysts were purchased from Sigma-Aldrich for use as a support in this work.

The incipient wetness-prepared DFMs were first characterized with XRD to ensure that the sorbent had made it onto the alumina support. The additional peaks around 18° and 20° 2θ in Figure 1 can likely be attributed to sodium oxide, indicating its presence after the incipient wetness preparation.³⁶ The largest peaks on the XRD (28°, 34°, 54°, and 67°) are consistent with an alumina diffraction pattern which has been used for the support of these DFMs.^37,38

Figure 1.

XRD pattern for Ru/alumina before and after NaO is added via incipient wetness.

Next, various concentrations of sodium oxide loadings were tested for the CO₂ capacity while preparing the DFM. DFM capacity was evaluated using CO₂ isotherms at 25 °C and partial pressures below ambient conditions. DFM with 3% sorbent loading was picked as a starting point, since it is the maximum amount that could be distributed on the support in one round of incipient wetness. Further rounds of incipient wetness were used to test higher loadings; however, these further rounds did little to increase the DFM CO₂ capacity. As seen in Figure 2, the DFM CO₂ capacity remained nearly constant when the sodium oxide loading was increased from 3 to 6%. Despite this lack of difference, 6% material had already been prepared, and further testing was under way, so it is used throughout the rest of the work presented here. It is likely that 3% loading would provide adequate CO₂ capture capacity based on these isotherms; however, they do not consider other real-world conditions like moisture content.

Figure 2.

Carbon dioxide adsorption isotherms at 25 °C for the pellet shaped alumina support with various sodium oxide loadings.

It is important to test potential CO₂ capture sorbents with the full mix of gases they will be exposed to during use, such as water, nitrogen, and oxygen. Water is a particularly interesting component in the air mixture, since it often interacts strongly with various CO₂ sorbents. For example, water will reduce 13X zeolites capacity for CO₂ down to almost zero as the zeolite will preferentially capture water over CO₂. Water and CO₂ coadsorption isotherms were measured in the DVS system to address this concern with results plotted in Figure 3.

Figure 3.

Low pressure carbon dioxide (normalized by dry mass or sorbent) isotherms for DFMs with various amounts of water preloaded onto the material (dry, equal parts, and saturated). Dry 13X zeolite is plotted in gray as a reference material for the state of the art in cabin air carbon dioxide capture.³⁹

Interestingly, water enhances the DFM capacity for CO₂. All of the wet DFM conditions tested showed increased CO₂ capacity compared to the dry DFM baseline (purple in Figure 3). The first set of wet conditions with 1:1 CO₂ to water began by dosing water over the material at low pressures (<3.17 kPa) at the same flow rate to be used for the CO₂ experiment. After dosing water over the material at low pressure, the CO₂ flow began so that further mass increases could be attributed to just CO₂. The amount of CO₂ adsorbed was normalized by the dry mass of the sorbent used. This scenario is plotted in green in Figure 3. The powder showed a slightly higher capacity than the pellet at all conditions; however, it is plotted only for the one 1:1 condition (in red). DVS does not allow for precise control of the inlet water stream content as it is set up, so it is hard to control a particular relative humidity. To test a further increase in the water content, the DFM was left to saturate with water overnight before the trial began (blue). This condition represents the highest water content that can be dosed in the DVS and shows increased capacity from the equal parts water and CO₂ condition. Improved capacity for similar DFM has been reported in the literature and attributed to the water forming carbonate complexes with the sodium sorbent. Similar capacity improvements have been seen by others in similar composition DFM for DAC.

Improved DFM performance with more realistic wet conditions also offers overall improvements to the CO₂ removal system by eliminating the need for a drying subsystem. The capacity of the wet DFM is higher than dry 13X zeolite at the same pressures, which represents a slight absolute CO₂ capacity improvement.³⁹ More importantly, DFM also represents a reduction in system complexity and size, since the air does not need to be dried before entering the CO₂ sorbent bed. Saving system mass, size, and complexity are all very important parameters to consider when evaluating flight CO₂ removal systems.⁴⁰

After considering the water/CO₂ binary system, CO₂ was captured from simulated cabin air at various flow rates (20–40 sccm) as seen in Figure 4. The sorbent (40 mg dry mass) was baked out under heating (350 °C) and vacuum prior to the start of the experiment. During pressurization, water was dosed into the air via the DVS system at 5 sccm, while air flowed in at the experimental flow rate. Water can only be dosed at or below its water pressure in the DVS (3.17 kPa at 25 °C), so very little water was dosed during pressurization (0.8% of gas or less). Once pressurized, simulated dry cabin air was allowed to flow through, and mass increase was measured as plotted in Figure 4.

Figure 4.

CO₂ adsorbed by DFM at different dry and wet simulated cabin air flow rates.

The DFM captured more CO₂ with water present for all tested air flows by up to nearly 50% at the lowest 20 sccm flow rate. Lower flows captured more CO₂ for both dry and wet cabin air, likely due to the increased residence time. DFM would likely capture more CO₂ as the flow further decreases until a maximum is reached. The flows tested are too high compared to the bed size to see any sort of maxima or improvements from higher flow over the sorbent. Initial capture rates (first 3 min) were also evaluated from the data in Figure 4. The capture rate was about 10× higher for the wet air compared to dry air at any given flow rate. The wet DFM capture performance is promising for DFM use with cabin air CO₂ capture, since cabin air will always be humid. Most CO₂ capture materials need a drying step prior to capture, just for the air to be rehumidified before being returned to the cabin. Eliminating this step to deal with water would decrease the CO₂ removal system complexity, size, mass, and power use.

After the study of the CO₂ capture capacities and rates, some initial work was done on converting the captured CO₂ to water. These studies utilized the DVS cell while a purpose built reactor was being built, so they are a bit limited in scope. After CO₂ was captured, like in the previous section, hydrogen flow began at 20 sccm (atm pressure) over the bed while it was heated at 5 °C/min. The reaction was monitored via FTIR at the outlet as seen in Figure 5. CO₂ is initially released from the sorbent under heating, as indicated by its early peak in the FTIR spectrum. It has been suggested that CO₂ can be conserved in this step by heating faster; however, the maximum heating rate of the DVS system was already in use. Once the temperature is high enough (∼300 °C), the CO₂ begins converting to the reaction products: methane and water. Hydrogen is not included in this plot since it is not FTIR-active. This initial test confirms the DFM used can convert the captured CO₂ into water and methane, as would be necessary for closed loop life support. More testing and experimental parameter exploration will be done in the dedicated reactor system once the construction is complete.

Figure 5.

Outlet stream composition via in line FTIR during the DFM reaction step when captured CO₂ is reacted with hydrogen to form water and methane.

After completing benchtop testing, the next steps will be to move DFM to a larger scale packed bed system for operation with a four member crew. There have been several packed bed systems like CDRA and 4BCO2 in the past which utilized 5A and 13X sorbents, respectively.^39,41 A full scale DFM system will work similarly to past hardware in that it will need multiple sorbent beds operating out of cycle to create a continuous process. However, no initial drying step will be needed before CO₂ capture due to DFM improved CO₂ capacity from moist air. Water would be separated from the product gas stream by condensation and centrifugation (in zero G scenarios), just like it was processed in the Sabatier reactor previously. Process models used previously for these systems were utilized to estimate the power required to heat up a full scale DFM bed to operating temperature (300 °C) and found that the DFM systems would represent around 250 W savings compared to the capture and conversion systems it would be replacing, as estimated by their respective heating requirements and shown in Table 1, while also cutting size and weight with the elimination of the water adsorbent beds. The numbers in Table 1 represent estimates for the power requirements of each system to go through its heating cycle, which is the main power draw for all of these temperature swing based systems. Additionally, these power savings represent a conservative estimate since they only consider power required for heating an adsorbent bed and do not account for any heating provided by the exothermic Sabatier reaction that would be occurring in the beds.

Table 1. Power Comparison of a Sorbent-Based Capture and Conversion System Assembly to a DFM System Based on Estimated Power Requirements for Heating.

system	estimated power (W)	estimated DFM power (W)
four-bed	385
CO2 compressor	230
Sabatier	314
total	929	675

4. Conclusion

DFMs were shown to be a promising technology for CO₂ capture and conversion to enhance recycling and close the loop in crewed spacecraft LSS. The Na₂O/Ru/Al₂O₃ DFM tested showed a CO₂ capture capacity from wet air comparable to the ISS state-of-the-art 13X zeolite from dry air. By having comparable performance with wet air to a dry zeolite sorbent, the DFM CO₂ removal system for LSS could eliminate the need for upstream drying beds. Additionally, wet DFM beds showed improved CO₂ uptake rates compared to dry feeds, which could allow for quicker adsorption cycle times. Eliminating the drying sorbent beds and combining CO₂ capture, pressurization, and conversion into one single set of beds would allow for significant mass (about half), volume (about a third), power (about 200 W or ∼22%), and complexity savings compared to the current air revitalization in use on ISS now. DFM is a promising technology for spacecraft LSS that could help enable longer duration human space missions, such as an established human presence on the moon or even a crewed trip to Mars.

Author Present Address

^§ West Biofuels, Woodland, CA 95776, USA

The authors declare no competing financial interest.