Why this class exists
Making Propellant and Water Off Earth Decides Whether Anything Stays
Every settlement and depot plan assumes local water and oxygen; very little has been made yet.
In-situ resource utilization turns local material into propellant, water, oxygen, metal and building stock, so it does not have to be launched. Off Earth, the total so far is 122 grams of oxygen from MOXIE on Mars, a lunar drill that ran for about 10 hours on a tipped lander, and samples and simulant bricks returned to Earth. No lunar polar ice has yet been measured at the surface: NASA's VIPER and China's Chang'e-7 both slipped to 2027. Observed
Extraction is not the expensive step. Electrolysis, which splits water into hydrogen and oxygen, costs about 6.3 kWh per kilogram of water at 70% cell efficiency, which is why ISRU and Space Nuclear power arrive together. The destination pages carry the site figures: lunar polar ice at about 4-6% by weight in NASA's 2025 review, Ceres crust ice, and Martian subsurface ice. Modeled
Demonstrated
Oxygen on Mars, Water in Chang'e-5 Soil, Bricks From Tiangong
What has actually worked, in grams and hours.
| Demonstration | Result | Detail | Rated |
|---|---|---|---|
| MOXIE · NASA, Perseverance | 122 g oxygen | 16 runs on Mars from CO₂, peak 12 g per hour at 98%+ purity; ended August 2023 | Observed |
| PRIME-1 · NASA on Intuitive Machines IM-2 | Drill and spectrometer worked | The lander tipped over in March 2025; about 10 hours of operation instead of 10 days, and no lunar ice detected | Observed |
| Chang'e-5 samples · CNSA | Water in returned regolith | Water confirmed from in-situ spectra and samples; a hydrated mineral, ULM-1, rich in water and ammonium (2024) | Observed |
| Regolith-simulant bricks · Tiangong | 74 bricks exposed | First batch returned November 2025 after one year outside; reported more than three times ordinary brick strength | Observed |
| Carbothermal oxygen reactor · NASA | TRL 6 | Reached about 1,800 °C in vacuum tests in 2025 | Observed |
Lunar
The South-Pole Ice Measurement Slipped to 2027 on Both Sides
VIPER and Chang'e-7 are the first missions built to measure polar ice where it lies.
| Mission · organization | Target | Status | Rated |
|---|---|---|---|
| VIPER · NASA, on Blue Moon MK1 | Mons Mouton, south pole | Cancelled 2024, reinstated September 2025; delivery on New Glenn in 2027, now exposed to the LC-36 rebuild | Building |
| Chang'e-7 · CNSA | Shackleton rim, 88.8°S | Postponed on 24 August 2026 to 2027; its hopper carries a water analyzer into shadowed craters | Building |
| Chang'e-8 · CNSA | Leibnitz Beta | Around 2028; 3D-prints bricks from regolith and carries a Pakistani rover | Projected |
| Prospect Moon · Interlune | Helium-3 extraction | $6.9M NASA award (May 2026) for a 2028 payload; 100 t per hour excavator prototype | Projected |
| Magna Petra | Helium-3 prospecting rover | 2028, reusing NASA's MSOLO spectrometer | Projected |
| Blue Alchemist · Blue Origin | Solar cells and oxygen from regolith | $35M NASA Tipping Point; critical design review September 2025 | Building |
| Pegasus Lunar Terrain Vehicle · Lunar Outpost | South-pole mobility | $220M NASA award 2 June 2026, delivery by 2028 | Building |
| PROSPECT · ESA | Drill and lab | Moved from the cancelled Russian Luna-27 to an Intuitive Machines Nova-C lander | Building |
Asteroids
China Is Sampling a Near-Earth Asteroid; US Startups Are Still Reaching One
Asteroid ISRU is a sampling problem today and a mining question later.
| Mission · organization | Target | Status | Rated |
|---|---|---|---|
| Tianwen-2 · CNSA | Kamo'oalewa, quasi-moon of Earth | Arrived mid-2026; sampling through April 2027, capsule return November 2027, then comet 311P | Observed |
| DeepSpace-2 · AstroForge | Near-Earth asteroid flyby | Rides IM-3, end of 2026; Odin was lost in 2025 | Building |
| High Frontier · Karman+ | Carbonaceous NEO sample | Slipped to 2027 | Projected |
| Asteroid capture · TransAstra | Small NEO | 10 m capture bag tested on the ISS October 2025; capture mission 2028 | Projected |
Contested
Deliberately Not Claimed
The load-bearing uncertainties, tagged with the confidence they actually earn.
Polar ice grade and form are unmeasured in place. Orbital data show hydrogen and cold traps; whether the ice is pore-filling, layered or frost decides whether extraction is mining or scraping.
Helium-3 has a funded payload and no proven demand. Interlune and Magna Petra fly in 2028. The large demand case rests on fusion reactors that do not exist; the near market is neutron detectors.
Lunar metal is halogen-gated. Every demonstrated route to lunar aluminum runs through halide chemistry, and lunar chlorine is tens of parts per million. Molten regolith electrolysis avoids it but yields alloy, not pure metal.
Record
Provenance and Data Sources
- Confidence
- Observed for flown, fielded or measured - Building for hardware and programs in development - Projected for announced plans - Modeled for original analysis
- Status date
- Q4 2026
- Mars and Moon
- JPL MOXIE results (2023); NASA PRIME-1 (March 2025); NASA lunar ISRU progress review (2025); VIPER reinstatement (September 2025)
- Commercial
- Interlune NASA award (May 2026); Blue Alchemist CDR (September 2025); Lunar Terrain Vehicle awards (June 2026); AstroForge, Karman+, TransAstra
- China
- CNSA Chang'e-7 postponement (24 August 2026); Chang'e-8 plan; Tianwen-2 arrival (2026); Tiangong brick return (November 2025)
- Weakest link
- No in-place measurement of polar ice; no off-Earth propellant plant has run