space · manufacturing · family: making one is easy. making a million is the problem
half the oxygenvented to space
The space station loses half its oxygen to space because no one can close the carbon loop
Problem statement
Astronauts exhale CO₂ that contains oxygen originally extracted from water, the station's chief source of oxygen. The ISS captures this CO₂ and reacts it with hydrogen in a Sabatier reactor to recover some oxygen as water, but the process produces methane as a byproduct, and the hydrogen bound up in that methane is not recovered — so there is not enough hydrogen to react all the CO₂, and about half of the CO₂ is vented. The result: the ISS's Carbon Dioxide Reduction Assembly recovers only ~47% of the oxygen from CO₂ (NASA's average estimate), and the rest of the oxygen the crew needs is shipped up from Earth. For missions to the Moon or Mars lasting months to years, resupply of oxygen becomes economically and logistically difficult — closing this loop is not an optimization, it is what NASA says will enable human journeys to Mars and beyond.
Why this matters
Open-loop life support treats water and oxygen as consumables, which NASA says brings considerable mass and logistics burdens on long missions beyond low Earth orbit; the ISS systems are only partially closed and require frequent resupply. NASA's SCOR project aims to raise oxygen recovery to more than 75%, with a stretch goal of 100% — more than double the current ~47% — to reduce the mass and volume of life-support consumables for Gateway, lunar and Mars missions. Past funding cuts and program discontinuations have created what researchers describe as "critical gaps" and a "strategic risk to US leadership in human space exploration" in bioregenerative life support.
What’s been tried and why it hasn’t worked
The Sabatier process (CO₂ + 4H₂ → CH₄ + 2H₂O) is the current ISS baseline, but its stoichiometry inherently limits oxygen recovery because the methane byproduct consumes hydrogen. The Bosch process (CO₂ + 2H₂ → C + 2H₂O) has a theoretical 100% oxygen recovery rate and produces solid carbon instead of methane, but the carbon deposits on the catalyst and accumulates in the reactor; NASA's SCOR page shows magnified fibers ringed with deposited carbon and a subscale reactor with accumulated carbon, where an attritor is used to mix the catalyst and keep the carbon particles from aggregating. Methane pyrolysis (CH₄ → C + 2H₂) could recover hydrogen from Sabatier methane; NASA's industry partner has operated it on a pilot test stand, and NASA's Life Support Systems Project expects to perform integrated testing and flight demonstrations for promising candidate technologies. Bioregenerative life support (plant-based systems) could in principle close the loop entirely but brings integration problems within a closed system — bio-safety, waste processing, water usage, bioprocess engineering — and the 2025 review says ground test facilities need to run for several years to understand reliability and failure issues. No nation has demonstrated a completely closed bioregenerative system that includes regenerative operation, feedstock and nutrient recycling, and human waste processing; the review cites China's Lunar Palace, which sustained a crew of four for a full year yet still did not close the loop on waste recycling, while NASA's own integrated testbed (BIO-Plex) was cancelled in 2004.
What would unlock progress
Near-term progress requires either: (1) a continuous Bosch reactor design that manages carbon deposition — NASA's SCOR project with Umpqua Research is pursuing this approach; or (2) reliable methane pyrolysis hardware that feeds recovered hydrogen back to the Sabatier, closing the hydrogen loop. Bioregenerative approaches need sustained investment in controlled-environment crop production research that was defunded in the 2000s. The integration challenge is as significant as the component challenge — individual subsystems must function reliably as a coupled system for years, and NASA's own framing is that the system must be robust, reliable and maintainable.
Entry points for student teams
A chemical engineering team could design and test a small-scale Bosch reactor prototype, focusing specifically on the carbon management problem: how to continuously extract deposited carbon from the catalyst bed without shutting down the reaction. This is a well-defined reaction engineering problem with clear success metrics (carbon accumulation rate, catalyst lifetime, reactor uptime). A biology/engineering team could set up a small closed-loop algae photobioreactor that takes in CO₂ at concentrations representative of cabin air (NASA-STD-3001 limits the one-hour average CO₂ partial pressure in the habitable volume to no more than 3 mmHg) and measures oxygen production rates, stability over weeks, and failure modes — directly informing the bioregenerative approach.
Genes
Structural cousins: stuck the same way, in other fields
Sources
"Environmental Control & Life Support System (ECLSS): Human-Centered Approach," NASA-STD-3001 Technical Brief OCHMO-TB-002 Rev A, NASA Office of the Chief Health & Medical Officer, April 20, 2023. (accessed 2026-02-14). Supplemented with "SpaceCraft Oxygen Recovery (SCOR)," NASA STMD Game Changing Development, project page last updated February 1, 2024 (project introduced October 2017). (accessed 2026-02-14). Also "Next Generation Life Support (NGLS)," NASA STMD Game Changing Development, December 2, 2020 (archived project page). (accessed 2026-02-14). Also D. Marshall Porterfield et al., "Critical investments in bioregenerative life support systems for bioastronautics and sustainable lunar exploration," npj Microgravity 11, 57 (2025). (accessed 2026-02-14). go to source 1 ↗ go to source 2 ↗ go to source 3 ↗ go to source 4 ↗
Working notes
Notes made while writing and checking this brief, shown as written. They use working shorthand.
- The `temporal:static` tag is strongly justified: the Sabatier process's ~47% oxygen recovery is limited by its own chemistry — hydrogen lost to the methane byproduct — and the ISS still runs it as the state of the art.
- The `failure:lab-to-field-gap` tag applies to the Bosch process — it has a theoretical maximum recovery of 100%, but a continuous reactor that manages the deposited carbon is still under development (a subscale reactor at NASA's partner), with flight demonstration still ahead.
- The `failure:not-attempted` tag applies to bioregenerative approaches — NASA's bioregenerative life support research was largely defunded after the Controlled Ecological Life Support System (CELSS) program ended, and the 2025 npj Microgravity review describes this as creating "critical gaps" and "strategic risk."
- Cross-domain connection: the carbon management challenge in the Bosch reactor shares structure with catalyst deactivation problems in industrial chemical processing — the reactor works until the catalyst is fouled, which is a common chemical engineering failure mode.
- Related briefs: `space-isru-regolith-oxygen-extraction` (both involve oxygen recovery from non-traditional sources in space); `critical-minerals-waste-extraction` (both involve extracting valuable materials from complex, contaminated feedstocks).
- NASA's SCOR project is partnered with Umpqua Research (Bosch) and Honeywell Aerospace (hydrogen recovery); the SCOR page lists the project as introduced in October 2017, while the NGLS page is marked as an archived project that is no longer updated.
Reconciliation 2026-09-30: Fact repair after reading the whole write-up against every source it names; a second reader went over the repairs on the same day, re-fetched all four sources, reworded four repairs and restored one original sentence. All four sources were fetched on 2026-09-30 and read in full: the NASA OCHMO technical brief (https://www.nasa.gov/wp-content/uploads/2023/07/eclss-technical-brief-ochmo.pdf, 24 pages, direct PDF), the SCOR project page (https://www.nasa.gov/spacecraft-oxygen-recovery-scor/, direct), the NGLS project page (https://www.nasa.gov/directorates/stmd/game-changing-development-program/next-generation-life-support-ngls/, direct; the page is marked archived and no longer updated), and the npj Microgravity paper (https://www.nature.com/articles/s41526-025-00518-4, direct, open access). (1) The text said methane is vented to space "carrying away half the hydrogen atoms needed for the next cycle." The NGLS page says that because methane is produced there is insufficient hydrogen to react all the carbon dioxide and about half of the CO₂ is vented; the SCOR page says the Sabatier requires additional hydrogen because of the methane byproduct. The sentence now says that. (2) "More than half the crew's oxygen supply is effectively lost overboard" and "this 53% loss rate" were arithmetic on the 47% figure; the SCOR page says the remaining oxygen the crew needs is transported from Earth, and the sentence now says that. "Mars missions lasting 2–3 years" is in no source; the OCHMO brief says missions to the Moon or Mars "lasting months to years," and "architecturally unacceptable" became the SCOR page's wording that resupply becomes economically and logistically difficult and that oxygen recovery will enable human journeys to Mars. "Extracted from water by electrolysis" now says extracted from water, the station's chief source of oxygen, as the SCOR page puts it. (3) Removed because no source contains them: launch costs of $2,700 (Falcon 9) to $54,000 (SLS) per kilogram; the six-person, 1,000-day Mars mission needing about 30,000 kg of water and oxygen; and 90% recovery saving thousands of kilograms. The paragraph now gives what the NGLS and SCOR pages say: open-loop systems carry considerable mass and logistics burdens, ISS systems are only partially closed and need frequent resupply, and SCOR aims for more than 75% recovery with a stretch goal of 100%, more than double the current figure. (4) The Bosch sentence said catalyst beds "clog" and "must be replaced," an "unacceptable consumables burden." The SCOR page shows magnified fibers ringed with deposited carbon and a subscale reactor with accumulated carbon and an attritor used to mix the catalyst and keep carbon particles from aggregating; it does not say the catalyst is replaced. The sentence now follows the page, and "without consumable catalyst replacement" was dropped from the Bosch item under What Would Unlock Progress for the same reason. (5) Methane pyrolysis "adds system complexity and has not been demonstrated in spaceflight-relevant hardware" is not in any source; the SCOR page shows Honeywell's pilot test stand in operation and says the Life Support Systems Project expects to perform integrated testing and flight demonstrations for promising candidate technologies. The sentence now says that. (6) "No bioregenerative ECLSS has been tested beyond laboratory-scale closed-chamber experiments lasting weeks" is contradicted by the npj paper, which says China's Lunar Palace sustained a crew of four for a full year and that BIOS-3 ran for nearly 30 years with human-subject experiments; the paper's own claim is that no nation has demonstrated a completely closed system that integrates regenerative operation, feedstock and nutrient recycling, and human waste processing, and that even Lunar Palace did not close the loop on waste recycling. The sentence now says that, and the list of bioregenerative difficulties (biological variability, lighting, crop disease, contamination) was replaced with the paper's list (bio-safety, waste processing, water usage, bioprocess engineering) and its statement that ground facilities need to run for several years to understand reliability and failure issues; "algae or" was dropped because no source mentions algae. (7) The sentence on dual-function materials replacing "CDRA + Sabatier + OGS" was removed; no source mentions dual-function materials, and the NGLS page calls the oxygen generator the OGA. (8) "Defunded in the 2000s" was left as written: the npj paper says BIO-Plex funding was terminated and the program cancelled in 2004, which supports it. (9) "For years without maintenance" is not in any source; the NGLS page asks for systems that are "robust, reliable, and maintainable," and the sentence now says that. (10) The entry point's "~0.5–1% CO₂" as representative of cabin air is in no source, and the OCHMO brief's requirement [V2 6004] limits the one-hour average CO₂ partial pressure to no more than 3 mmHg; the parenthesis now cites that limit. (11) In the notes above, "understood since the 1890s" and "activated in 2010" are in no source and were removed; the Bosch note's "achieves near-100% recovery in laboratory batch reactors" now says theoretical maximum of 100%, which is the NGLS page's wording; and "active investment priority" now reports what the pages say (SCOR introduced October 2017; NGLS archived). (12) The Source line now carries the titles, document number and dates as they appear on the documents: the OCHMO brief is OCHMO-TB-002 Rev A, dated April 20, 2023; the SCOR page carries a last-updated date of February 1, 2024 (which the earlier "2024" reflected) and lists the project as introduced October 2017; the NGLS page is dated December 2, 2020; and the npj paper's full title, first author (D. Marshall Porterfield et al.), volume 11 and article number 57 were copied from the publisher record. Confirmed in the sources and left as written: the Carbon Dioxide Reduction Assembly's average estimated recovery of about 47%; about half of the CO₂ vented; Sabatier producing water and methane from CO₂ and hydrogen; Bosch producing solid carbon and water with a theoretical maximum recovery of 100%; methane pyrolysis proposed to recover hydrogen; SCOR's Continuous Bosch Reactor with Umpqua Research and hydrogen recovery with Honeywell Aerospace; the quotations "critical gaps" and "strategic risk to US leadership in human space exploration," which appear verbatim in the npj abstract; "funding cuts and program discontinuations"; and the recommendation for renewed program investment. Not in any source but left as written because they are textbook chemistry rather than source claims: the balanced equations for the Sabatier, Bosch and methane pyrolysis reactions. No title changed: "half" matches the NGLS page's "about half is vented," though the sources say half of the oxygen in exhaled CO₂, not half of the station's oxygen. Genome tags untouched.