Lunar ISRU is the difference between a short expedition and a self-sustaining Ark. The strongest near-term paths are water extraction from polar permanently shadowed regions, oxygen extraction from regolith, and regolith-based construction; metal production and full propellant chains remain earlier-stage but are advancing quickly.
Executive assessment
- Water ice is the highest-value lunar resource because it supports life support, radiation-shielded habitat operations, and propellant production.
- Oxygen extraction from regolith is the most mature bulk-material ISRU pathway today, with multiple successful demonstrations.
- Construction from regolith is already partially demonstrated in printing and sintering tests, but not yet at habitat-scale deployment.
- Aluminium and titanium smelting from lunar soil remains mostly theoretical or lab-scale, with MRE and related processes proving oxygen and leaving metal-rich slag as a useful byproduct.
- Propellant production is not yet demonstrated end-to-end on the Moon, but the feedstock chain is clear: water extraction → electrolysis → hydrogen/oxygen propellant.
1) Water ice extraction from polar permanently shadowed regions
Lunar water ice is concentrated in permanently shadowed regions, especially near the poles, where temperatures can remain below about 75 K in the coldest traps.[1] NASA states that evidence for water ice is widespread in PSRs and that greatest concentrations are expected in the coldest locations and on poleward-facing slopes.[1]
Specific remote-sensing results in the record include estimates of 1.5 ± 0.8 wt.% water-equivalent hydrogen in polar regions if confined to PSRs, and 0.3–0.5 wt.% in the strongest neutron-signature areas.[1] A 2025 analysis reported 1,578 water-ice-containing pixels in lunar shallow subsurface data, with 1,445 pixels, or about 91%, clustered in 29 PSRs.[1]
What has been demonstrated:
- Surface-exposed water ice has been directly observed in lunar polar regions.[1]
- Orbital datasets have repeatedly identified PSRs as the prime targets for extraction planning.[1]
What remains unproven:
- No operational polar ice mining system has yet been demonstrated on the Moon.
- The engineering problem is not detection; it is excavation, thermal control, volatile capture, and contamination-free processing in extreme cold and darkness.
Mission implication:
- For the Ark, ice extraction is the highest-leverage first industrial system.
- Design targets should assume hundreds of grams to kilograms per hour only after pilot validation; current evidence is still below production engineering.
2) Oxygen extraction from regolith
This is the most advanced lunar resource-processing category.
NASA’s 2026 Lunar Surface Technology update states that carbothermal reduction is the highest-TRL oxygen-extraction approach from regolith and that NASA’s molten regolith electrolysis (MRE) has been advanced through NASA partnership work.[2] The same source says MRE has been successfully tested in a vacuum environment with NASA gas analysis hardware, demonstrating oxygen extraction from regolith simulant and leaving metal-rich slag as a byproduct.[2]
The strongest quantified demonstration in the record is the late-2024 KSC/Lunar Resources LR-1 test: it processed 25 kg of regolith simulant, ran under vacuum, and achieved an average oxygen production rate of 0.07 kg/hr over 9 hours of sustained electrolysis.[1] A NASA progress review also reports that the MRE reactor processed 25 kg of lunar highland simulant over 36 hours with 24 hours of electrolysis, and that measured oxygen production matched theoretical levels.[3]
Carbothermal reduction has also shown meaningful performance: NASA’s review reports >20 g O₂/kW-hr thermal, oxygen yields of >20% by mass relative to regolith, and >99.7% recovery of carbon used in melt tests.[3] The same review describes single-melt demonstrations equivalent to 140 kg O₂/year production rate.[3]
What has been demonstrated:
- Oxygen extraction from simulants in relevant vacuum environments.[1][2][3]
- Sustained MRE operation at meaningful batch scale: 25 kg feedstock, 0.07 kg/hr average oxygen output.[1]
- Carbothermal reduction reaching strong efficiency metrics and near-complete carbon recovery.[3]
What remains theoretical or incomplete:
- Continuous, autonomous, fieldable lunar oxygen plants have not yet been proven.
- Feedstock handling, electrode life, thermal cycling, and contamination management still need lunar validation.
- Conversion from oxygen production to reliable life-support and propellant-scale throughput remains unproven.
Mission implication:
- Oxygen from regolith is the Ark’s industrial backbone for breathing gas, oxidizer, and internal manufacturing.
- The target is not scientific novelty; it is steady-state tonnage.
3) Aluminium and titanium smelting from lunar soil
Direct extraction of useful metals from regolith is the least mature of the major ISRU pathways, but it is strategically important because the byproducts of oxygen extraction are metal-rich residues.
NASA explicitly notes that MRE produces metal-rich slag that can be processed for manufacturing.[2] That makes oxygen extraction a precursor to metal recovery, not a competitor to it. In practice, lunar soil contains abundant aluminium-bearing silicates and titanium-bearing ilmenite, but refining them into structural metal requires high-temperature, high-purity, high-reliability metallurgy that has not yet been demonstrated in lunar conditions.
What has been demonstrated:
- Metal-rich slag generation from oxygen extraction processes.[2]
- Early-stage process integration studies for using lunar regolith as both oxygen feedstock and structural byproduct source.
What is still mostly theoretical:
- Full aluminium smelting from lunar soil at industrial scale.
- Full titanium extraction and refining from lunar soil at industrial scale.
- Closed-loop, lunar-environment metallurgical refining chains.
Mission implication:
- Do not plan on primary lunar structures made from smelted Al/Ti in the first industrial wave.
- Plan on