
- Key Takeaways
- Blue Moon’s Refueling Need and Working Assumptions
- How Much Water Ice a Blue Moon Refill Would Require
- Why Oxygen Comes Out in Surplus
- Mining Mass and the Lunar Ice Grade Problem
- Energy Needed to Extract and Convert Water Ice
- Timelines for a Single Refill Campaign
- Power Architecture for Polar Propellant Production
- Operational Risks for a Blue Moon Fuel Depot
- Summary
- Appendix: Useful Books Available on Amazon
- Appendix: Top Questions Answered in This Article
- Appendix: Glossary of Key Terms
Key Takeaways
- A 40-metric-ton refill may require about 51 metric tons of lunar water.
- Electrolysis dominates energy demand before hydrogen liquefaction and ice extraction.
- Ice grade, plant scale, and polar power access drive the production timeline.
Blue Moon’s Refueling Need and Working Assumptions
NASA selected Blue Origin on May 19, 2023, to develop the Blue Moon lander for the agency’s Artemis human landing system program, with a $3.4 billion firm-fixed-price award that included an uncrewed demonstration mission before a crewed demonstration under the Artemis V landing architecture described at the time. In March 2026, NASA updated the Artemis transportation sequence, adding a 2027 low Earth orbit demonstration mission to test system capabilities closer to Earth before a planned 2028 astronaut mission to the lunar South Pole. As of May 19, 2026, NASA still describes Blue Origin as one of the two American companies developing human landing systems to move astronauts from lunar orbit to the lunar surface and back for Artemis.
Blue Origin has not publicly released a full propellant loading specification for the crewed Blue Moon Mark 2 human lunar lander. That absence matters because any estimate of lunar water ice needs must start with the amount of liquid oxygen and liquid hydrogen that must be loaded into the lander. Publicly available information confirms the basic architecture: Blue Origin’s Blue Moon family includes the Mark 1 cargo lander and the Mark 2 crew and cargo landers, and the company’s BE-7 engine burns liquid oxygen and liquid hydrogen. Blue Origin lists the BE-7 at 10,000 pounds-force, or 44.5 kilonewtons, with deep throttling to 2,000 pounds-force, or 8.9 kilonewtons.
A practical estimate can still be built by treating Blue Moon as a reusable lunar lander that needs enough propellant for descent, ascent, maneuvering, hover margin, docking, reserves, and boiloff replacement. The working range used here is 30 to 50 metric tons of total loaded propellant, with 40 metric tons as the central case. This is not a published Blue Origin tank capacity. It is a sizing estimate consistent with a large hydrogen-oxygen lunar lander serving NASA’s sustained surface mission architecture, not a small robotic lander.
The central estimate assumes an oxygen-to-hydrogen mixture ratio of 6:1 by mass. That means 40 metric tons of total propellant would contain about 34.3 metric tons of liquid oxygen and 5.7 metric tons of liquid hydrogen. A lower 30-metric-ton case would contain about 25.7 metric tons of oxygen and 4.3 metric tons of hydrogen. A higher 50-metric-ton case would contain about 42.9 metric tons of oxygen and 7.1 metric tons of hydrogen.
This estimate treats “refill” as the usable propellant load placed into the lander’s tanks, not the total water mined. The distinction is necessary because water naturally splits into hydrogen and oxygen at an 8:1 oxygen-to-hydrogen mass ratio, but hydrogen-oxygen rocket engines usually run hydrogen-rich compared with the chemistry of water. If the lander consumes propellant at a 6:1 oxygen-to-hydrogen ratio, lunar water production creates excess oxygen relative to the hydrogen needed for the refill. That extra oxygen is not waste if it can feed life support, fuel cells, other spacecraft, surface mobility systems, or a separate oxygen-only customer.
The table below shows the working refill cases used for the article’s energy and mining estimates.
| Refill Case | Total Propellant | Liquid Oxygen | Liquid Hydrogen | Water Required | Excess Oxygen |
|---|---|---|---|---|---|
| Low Case | 30 Metric Tons | 25.7 Metric Tons | 4.3 Metric Tons | 38.6 Metric Tons | 8.6 Metric Tons |
| Central Case | 40 Metric Tons | 34.3 Metric Tons | 5.7 Metric Tons | 51.4 Metric Tons | 11.4 Metric Tons |
| High Case | 50 Metric Tons | 42.9 Metric Tons | 7.1 Metric Tons | 64.3 Metric Tons | 14.3 Metric Tons |
A 40-metric-ton Blue Moon refill, using the assumptions above, requires about 51.4 metric tons of water before losses. That figure does not include water lost during mining, vapor capture, purification, transfer, startup, shutdown, venting, leakage, or failed processing runs. A real production plan would add a margin. A 10% process loss raises the central water target to about 57 metric tons. A 25% loss raises it to about 69 metric tons.
How Much Water Ice a Blue Moon Refill Would Require
Water looks attractive for lunar propellant because it already contains both required chemical elements. By mass, water contains about one part hydrogen for every eight parts oxygen. Producing 1 kilogram of hydrogen from water consumes about 9 kilograms of water and produces about 8 kilograms of oxygen. That mass relationship drives the entire Blue Moon refueling problem more than any individual mining machine or power system.
For a 40-metric-ton propellant load at a 6:1 oxygen-to-hydrogen ratio, the lander needs about 5.7 metric tons of hydrogen. Producing that hydrogen from water requires about 51.4 metric tons of water. The same water creates about 45.7 metric tons of oxygen. Since the lander needs only about 34.3 metric tons of oxygen in this case, the plant would have about 11.4 metric tons of excess oxygen after filling the lander.
This is one reason hydrogen-oxygen propellant production has a different resource profile from oxygen-only extraction from lunar regolith. Oxygen accounts for most of a hydrogen-oxygen propellant load, but hydrogen is the limiting element when the source feedstock is water and the engine mixture is hydrogen-rich. Oxygen from water is abundant enough that surplus oxygen becomes part of the business case. A lunar surface installation could use surplus oxygen for astronaut breathing gas, oxidizer for surface hoppers, emergency reserves, or export to an orbital depot.
The amount of raw ice-bearing material depends on ice grade. The Lunar Crater Observation and Sensing Satellite impact site in Cabeus crater produced an estimated 5.6% plus or minus 2.9% water-ice concentration by mass in the excavated regolith, according to the published Science paper indexed by the NASA Astrophysics Data System. Lunar polar ice distribution remains uneven and insufficiently characterized for industrial planning. NASA’s broader lunar water summary notes that the Lunar Reconnaissance Orbiter and LCROSS findings support the presence of water ice in permanently shadowed regions, and later work confirmed water ice at multiple polar locations.
If a production site averaged 5.6% recoverable water by mass, a central 51.4-metric-ton water requirement would imply excavating and processing about 918 metric tons of icy regolith before losses. At 3% water by mass, the same refill would require about 1,714 metric tons of material. At 1% water by mass, it would require about 5,143 metric tons. These numbers explain why prospecting matters as much as electrolysis efficiency. A plant in low-grade material becomes a mining operation first and a chemical plant second.
The table below compares the mining mass implied by different ice grades for the three refill cases.
| Recoverable Water Grade | 30-Ton Refill | 40-Ton Refill | 50-Ton Refill | Operational Meaning |
|---|---|---|---|---|
| 1% | 3,857 Metric Tons | 5,143 Metric Tons | 6,429 Metric Tons | Large Mining Campaign |
| 3% | 1,286 Metric Tons | 1,714 Metric Tons | 2,143 Metric Tons | Heavy Robotic Excavation |
| 5.6% | 689 Metric Tons | 918 Metric Tons | 1,148 Metric Tons | Central Planning Case |
| 10% | 386 Metric Tons | 514 Metric Tons | 643 Metric Tons | High-Grade Site |
The table assumes all water in the processed material can be recovered. Field performance will be lower. The 2026 Lunar Water Extraction and Purification Technologies project demonstrated water extraction and capture under simulated permanently shadowed region conditions, recovering roughly half to 70% of the initial water in multi-kilogram tests, with a maximum near 73%. That work also found that vapor capture capacity and contaminants can restrict practical recovery.
Why Oxygen Comes Out in Surplus
Hydrogen-oxygen propulsion creates a counterintuitive lunar resource issue: the lighter propellant component controls the water requirement. Liquid hydrogen has excellent performance, but its low density, very low storage temperature, and high specific water demand make it the hard side of the production chain. Liquid oxygen is heavy in the tank, but it is chemically abundant in water and in oxygen-bearing lunar minerals.
For the central case, each 40 metric tons of lander propellant requires about 5.7 metric tons of hydrogen. That hydrogen fixes the water requirement at about 51.4 metric tons. The oxygen from that water exceeds the oxygen needed by the lander by about 11.4 metric tons. If the actual BE-7 operating mixture ratio differs from 6:1, the surplus changes. A higher oxygen-to-hydrogen ratio reduces excess oxygen. A lower ratio increases it.
This surplus oxygen affects both plant economics and architecture. If the plant discards excess oxygen, the energy and mining system serves a single lander refill less efficiently. If the plant stores excess oxygen, the same water-mining campaign supports a broader lunar propellant market. That market could include oxygen for crewed habitats, pressurized rovers, fuel-cell power systems, emergency reserves, oxygen-rich ascent stages, and oxygen transfer to a cislunar depot.
Oxygen’s storage burden is smaller than hydrogen’s. Liquid oxygen boils at about 90 kelvin, warmer than liquid hydrogen’s roughly 20 kelvin condition. The cryogenic challenge remains difficult on the lunar surface, but liquid oxygen is much easier to store than liquid hydrogen. NASA has treated cryogenic fluid management as an important capability area because long-term storage, transfer, and boiloff control are needed for modern lunar lander architectures. NASA’s March 2026 Artemis architecture update also points to docking, life support, communications, propulsion, and suit testing as part of the path toward later lunar landings, which makes fluid transfer and integrated system checks part of the wider lander readiness picture.
The oxygen surplus also offers a risk-control benefit. Early lunar propellant plants may first prove oxygen production and storage before full hydrogen liquefaction reaches operational scale. Oxygen extracted from water or regolith can serve crew systems and robotic vehicles even if liquid hydrogen storage is delayed. This staged approach would reduce pressure on the most difficult part of the process without abandoning the longer-term goal of refilling hydrogen-oxygen landers.
Mining Mass and the Lunar Ice Grade Problem
The most expensive kilogram in a lunar propellant plant may be the kilogram of water that is not found where expected. Orbital data confirm that polar water exists, but mission planners still need ground truth on depth, concentration, grain size, contaminants, mechanical properties, bearing strength, trafficability, and local terrain. A promising orbital detection does not automatically translate into a mineable reserve.
Permanently shadowed regions create the best cold traps but the harshest operating sites. UCLA’s Diviner Lunar Radiometer Experiment science team reports that temperatures in some permanently shadowed craters can fall as low as 25 kelvin, or about minus 414 degrees Fahrenheit. Nearby crater rims may receive long periods of sunlight and remain near 220 kelvin, making them more attractive for power systems, communications, thermal control, and human operations.
That split creates a basic engineering choice. The plant can place mining hardware inside the cold, dark ice area and route power into it, or it can move icy material out to a warmer processing area. Mining inside a permanently shadowed region reduces hauling distance but burdens every machine with darkness, severe cold, slow thermal recovery, and limited direct solar power. Hauling regolith to a sunlit or semi-sunlit processing area can simplify power and maintenance but adds robotic traffic, mechanical wear, and handling losses.
The central 40-metric-ton refill case gives useful scale. At 5.6% recoverable water, about 918 metric tons of icy regolith must be processed before losses. If a small robotic mining system processed 5 metric tons of regolith per day, the excavation phase alone would take about 184 operating days. At 20 metric tons per day, excavation would take about 46 operating days. At 50 metric tons per day, it would take about 18 operating days. Those figures exclude downtime, grading, route clearing, sample verification, maintenance, thermal cycling, storage tank conditioning, and propellant transfer.
At 1% recoverable water, the same refill grows into a much heavier campaign. The plant must process more than 5,000 metric tons of material before losses. Even at 50 metric tons per day, excavation would take more than 100 operating days. A low-grade deposit can still be valuable if it is shallow, accessible, and extensive, but the mining system must be sized accordingly. A high-grade deposit that is steep, blocky, or hard to reach may be less useful than a lower-grade deposit on safer terrain.
Contaminants add another constraint. Lunar polar volatiles may include substances beyond water, and extraction tests show that volatile contaminants can interfere with capture surfaces and reduce recovery. A propellant plant must purify water before electrolysis, dry and clean oxygen and hydrogen streams, and protect cryogenic equipment from dust. This is a space mining problem, a chemical processing problem, and a cryogenic logistics problem at the same time.
Energy Needed to Extract and Convert Water Ice
Electrolysis sets the main electrical demand after water reaches the processing plant. The United States Department of Energy’s 2024 proton exchange membrane electrolyzer cost record uses 55.2 kilowatt-hours per kilogram of hydrogen at beginning of life and an average system electricity consumption of 57.5 kilowatt-hours per kilogram across the modeled system life. For lunar planning, a 55 to 65 kilowatt-hour-per-kilogram range is a reasonable planning band for hydrogen production, allowing for balance-of-plant loads and off-nominal operation.
For the central 40-metric-ton refill, the lander needs about 5,714 kilograms of hydrogen. At 55 kilowatt-hours per kilogram, electrolysis alone requires about 314 megawatt-hours. At 65 kilowatt-hours per kilogram, it requires about 371 megawatt-hours. The oxygen is produced during the same electrolysis process, so these figures cover the chemical splitting of water into both gases, not a separate oxygen-making step.
Hydrogen liquefaction adds a second large load. The Department of Energy’s hydrogen liquefaction record states that industrial liquefiers use about 10 to 20 kilowatt-hours per kilogram of hydrogen, compared with a theoretical minimum of 2.88 kilowatt-hours per kilogram under the assumptions in that record. For the central Blue Moon case, that adds roughly 57 to 114 megawatt-hours.
Oxygen liquefaction is smaller in energy terms but still important. A planning band of 0.3 to 0.7 kilowatt-hours per kilogram of liquid oxygen places the central case at about 10 to 24 megawatt-hours. Oxygen liquefaction system performance in a lunar plant may differ from Earth industrial plants because it cannot use the same atmosphere, scale, maintenance regime, or heat rejection environment. The figure should be treated as a scoping range, not a final engineering value.
Ice extraction can dominate the total if the process is inefficient or the deposit is poor. A 2023 review of lunar in-situ resource use reported water extraction energy estimates of 8.6 to 37.9 watt-hours per gram, equal to 8.6 to 37.9 kilowatt-hours per kilogram of recovered water, depending on extraction approach and conditions. The 2026 LUWEX experiment also produced low water recovery rates in kilogram-scale tests, underscoring the gap between laboratory proof and industrial production.
The table below gives a scoping energy profile for the three refill cases. The low end combines favorable extraction and efficient processing. The high end includes more demanding extraction and less efficient cryogenic processing.
| Refill Case | Electrolysis Energy | Hydrogen Liquefaction | Oxygen Liquefaction | Ice Extraction | Total Energy Range |
|---|---|---|---|---|---|
| Low Case | 236-279 MWh | 43-86 MWh | 8-18 MWh | 332-1,462 MWh | 618-1,844 MWh |
| Central Case | 314-371 MWh | 57-114 MWh | 10-24 MWh | 442-1,949 MWh | 824-2,459 MWh |
| High Case | 393-464 MWh | 71-143 MWh | 13-30 MWh | 553-2,436 MWh | 1,030-3,074 MWh |
The central case lands near 824 to 2,459 megawatt-hours before transfer losses, boiloff losses, storage conditioning, construction energy, maintenance energy, and contingency reserves. A planning value near 1,200 to 1,800 megawatt-hours for one 40-metric-ton refill is a defensible middle range if the site has moderate water grade and the extraction method performs far better than early laboratory hardware but short of Earth industrial efficiency.
Timelines for a Single Refill Campaign
A single Blue Moon refill depends on average power, not just total energy. A plant that needs 1,500 megawatt-hours can finish the energy-intensive work in about 150 days at 10 kilowatts, 15 days at 100 kilowatts, or 1.5 days at 1 megawatt if every subsystem can operate at that rate. Real hardware cannot translate nameplate electrical power into finished propellant with perfect uptime, so calendar time will be longer.
A 100-kilowatt lunar propellant plant would be a large surface power installation by early Artemis standards. At that level, the central case energy range of 824 to 2,459 megawatt-hours implies 343 to 1,025 days of continuous operation. At 250 kilowatts, the same range becomes about 137 to 410 days. At 500 kilowatts, it becomes about 69 to 205 days. At 1 megawatt, it becomes about 34 to 102 days. These figures assume round-the-clock load availability and no major shutdowns.
Mining throughput imposes a second clock. In the central 5.6% water-grade case, roughly 918 metric tons of icy regolith must be processed before losses. A mining and handling system that processes 20 metric tons per day would need about 46 operating days just to move enough feedstock. If recovery is 70%, the required feedstock rises to roughly 1,312 metric tons, and the excavation period at the same rate becomes about 66 operating days. At 5 metric tons per day, the excavation period becomes many months.
The production sequence has bottlenecks that do not overlap perfectly. Site preparation, route mapping, robotic commissioning, regolith excavation, volatile extraction, water cleanup, water storage, electrolysis, gas drying, liquefaction, propellant storage, tank conditioning, lander transfer, and quality verification each constrain the schedule. Some tasks can overlap. Others must wait for clean feedstock, cold tanks, stable power, or lander arrival.
A near-term demonstration plant would probably operate at kilogram-per-day to tens-of-kilograms-per-day water scale. A lander-refill plant needs tons to tens of tons per month. That difference is why a first Blue Moon mission is unlikely to depend on lunar-made propellant. NASA’s Blue Origin contract path focuses on lander development, testing, verification, uncrewed demonstration, and crewed demonstration, not on making lunar surface propellant mission-essential for the first Blue Moon crewed landing.
A credible operational timeline for one central-case refill using a mature pilot-scale plant is likely measured in months, not days. With a 250 to 500 kilowatt average electrical supply, moderate ice grade, and a mining system sized for 20 to 50 metric tons of regolith per day, a 40-metric-ton refill might take roughly three to nine months after the plant is commissioned. With 1 megawatt of reliable power and a high-grade site, the same refill could fall toward one to four months. With low-grade ice, small excavation machines, or low power, the campaign could exceed a year.
Power Architecture for Polar Propellant Production
Power supply controls the practical refill rate. Lunar polar sites offer a rare advantage: some elevated areas near the poles receive long-duration sunlight compared with equatorial regions, and some lie near permanently shadowed regions that may contain ice. The problem is that the best sunlight and the best cold-trapped volatiles are often separated by terrain, shadow, slope, and line-of-sight constraints.
A surface propellant plant could use solar arrays on illuminated ridges, power cables or microwave links into colder mining zones, and batteries or regenerative fuel cells to ride through interruptions. Nuclear fission power would reduce dependence on local lighting, but it adds launch mass, policy review, safety analysis, and development complexity. NASA has continued to fund fission surface power work because a compact nuclear power system could support operations during darkness and in shadowed terrain. A hybrid architecture may fit early operations best: solar power for high-average output, stored energy for transient loads, and a smaller backup source for survival and safe shutdown.
The energy range in the central estimate shows why tens of kilowatts are not enough for regular lander refilling. At 50 kilowatts average power, 1,500 megawatt-hours requires more than three years of continuous operation. At 250 kilowatts, it requires about 250 days. At 1 megawatt, it requires about 63 days. Those figures cover energy only, before mining pace and maintenance downtime. A lunar propellant plant that supports recurring human lander refills needs industrial-class surface power.
| Average Plant Power | 824 MWh Low-Energy Case | 1,500 MWh Middle Case | 2,459 MWh High-Energy Case | Operational Interpretation |
|---|---|---|---|---|
| 50 kW | 687 Days | 1,250 Days | 2,049 Days | Demo Scale |
| 100 kW | 343 Days | 625 Days | 1,025 Days | Slow Refill |
| 250 kW | 137 Days | 250 Days | 410 Days | Pilot Industrial Scale |
| 500 kW | 69 Days | 125 Days | 205 Days | Early Operational Scale |
| 1 MW | 34 Days | 63 Days | 102 Days | Recurring Refill Scale |
The plant’s average power must be lower than or equal to the power that can be delivered through the full chain. A 1-megawatt solar field is not a 1-megawatt propellant plant if part of the power charges batteries, warms mechanisms, runs communications, operates rovers, rejects heat, and protects cryogenic stores. Average delivered processing power is the number that matters.
Thermal management has a double burden. Ice extraction requires heat in extremely cold areas, but liquid hydrogen storage requires keeping product near 20 kelvin. A poorly integrated plant might spend energy heating regolith and then spend more energy rejecting heat near cryogenic tanks. A better design physically separates hot extraction zones, water handling, electrolysis, gas storage, liquefaction, and cryogenic tank farms.
Operational Risks for a Blue Moon Fuel Depot
The largest risk is resource uncertainty. The Moon’s polar ice is real, but the grade, thickness, lateral continuity, contaminants, and mining behavior at a specific operating site remain open until robotic prospectors and drills measure them directly. NASA’s lunar water summary emphasizes that many questions remain about the origin, behavior, and future of lunar water even after multiple confirmations from missions and instruments.
The second risk is mechanical reliability in permanently shadowed terrain. Machines operating near 25 to 100 kelvin face brittle materials, lubricant limits, poor visibility, abrasive dust, and delayed repairs. Human crews are unlikely to spend routine work periods inside deep cold traps unless mission systems make that safe and efficient. Robotic mining must handle uncertainty without constant human intervention.
Hydrogen storage is the third risk. Liquid hydrogen is hard to store because its temperature is far below liquid oxygen and because hydrogen molecules are small enough to challenge seals and materials. Earth-based hydrogen liquefaction plants depend on large scale, steady operations, maintenance access, and known supply chains. A lunar liquefier must run with low maintenance, limited spares, no local industrial base, and high launch cost for every replacement part.
The fourth risk is operational coupling. A delay in mining can starve electrolysis. A cold-trap problem can waste water vapor. A liquefier fault can force gas storage or venting. A tank boiloff issue can erase production gains. A lander schedule slip can leave propellant sitting longer than intended. A power outage can force a safe-mode sequence that protects hydrogen, oxygen, water, electronics, and robotic vehicles at the same time.
Defense and security considerations will also shape lunar propellant infrastructure. A surface plant able to produce tens of metric tons of oxygen and hydrogen would be strategic infrastructure in cislunar space. It would support civil exploration, commercial logistics, emergency rescue planning, and potentially defense-related mobility or surveillance support. Governance questions under the Outer Space Treaty, Artemis Accords, national licensing, safety zones, spectrum use, and responsible surface operations would sit alongside the engineering problem.
The most realistic path is staged capability. Early missions verify ice deposits, test extraction, demonstrate water cleanup, produce oxygen, store cryogenic fluids, and transfer small quantities. Larger plants then scale water throughput, add hydrogen liquefaction, and connect to lander operations. A Blue Moon refill from lunar water ice is technically plausible in principle, but it requires a surface industrial base that goes well beyond a single lander or a single rover.
Summary
A 40-metric-ton refill for a Blue Origin human lunar lander is a reasonable central planning case for examining lunar water ice propellant, but it should not be read as a published Blue Origin specification. Under a 6:1 oxygen-to-hydrogen mixture assumption, that refill needs about 34.3 metric tons of liquid oxygen, 5.7 metric tons of liquid hydrogen, and 51.4 metric tons of water before losses. The same water would produce about 11.4 metric tons of surplus oxygen, which could become a valuable product for life support, power systems, other vehicles, or depot reserves.
The mining burden depends on ice grade more than any single chemical-process efficiency. At 5.6% recoverable water by mass, the central case requires about 918 metric tons of icy regolith before losses. At 1%, it requires more than 5,000 metric tons. That difference changes the project from a compact pilot operation into a large robotic mining campaign.
The energy burden is also large. A central refill requires hundreds of megawatt-hours for electrolysis and hydrogen liquefaction, plus extraction energy that could push the full campaign into the 824 to 2,459 megawatt-hour range. A middle planning value near 1,500 megawatt-hours is useful for scoping. At 250 kilowatts of average delivered plant power, that is roughly 250 days of energy production. At 1 megawatt, it is roughly 63 days before downtime and mining-rate limits.
The strongest near-term conclusion is not that lunar propellant is unavailable. The stronger conclusion is that the first practical Blue Moon refills from lunar water ice would require a purpose-built mining, power, processing, cryogenic storage, and transfer installation. The lander is one customer. The plant and moon base are the larger customers.
Appendix: Useful Books Available on Amazon
- The Value of the Moon
- Moon Rush: The New Space Race
- The Case for Space
- The Moon: Resources, Future Development and Settlement
- Lunar Sourcebook
- Resources of Near-Earth Space
Appendix: Top Questions Answered in This Article
How much water ice would be needed to refill a Blue Origin human lunar lander?
A central 40-metric-ton propellant refill would require about 51.4 metric tons of water before process losses, assuming a 6:1 liquid oxygen to liquid hydrogen mixture ratio. With 10% to 25% losses, the practical mining target rises to about 57 to 69 metric tons of recovered water.
Why does the estimate require more water than propellant?
Water contains oxygen and hydrogen at an 8:1 mass ratio. A hydrogen-oxygen rocket engine commonly uses a more hydrogen-rich mixture than that ratio. The hydrogen requirement drives the water demand, creating surplus oxygen after the lander’s liquid oxygen tank is filled.
How much icy regolith must be mined for one refill?
At 5.6% recoverable water by mass, the central refill case requires about 918 metric tons of icy regolith before losses. At 3%, the same case requires about 1,714 metric tons. At 1%, it requires more than 5,000 metric tons, making ice grade one of the largest drivers.
How much energy would one refill require?
The central case requires roughly 824 to 2,459 megawatt-hours across ice extraction, electrolysis, hydrogen liquefaction, and oxygen liquefaction. A practical middle planning value near 1,500 megawatt-hours is reasonable for early scoping, but site conditions and hardware performance could shift the total.
Which step uses the most energy?
Ice extraction and electrolysis are the largest energy loads in most cases. Electrolysis is predictable once clean water reaches the plant. Ice extraction is less certain because energy use depends on ice grade, regolith temperature, depth, contaminants, capture efficiency, and thermal losses.
Could surplus oxygen be useful?
Yes. Surplus oxygen could support crew breathing gas, fuel cells, surface vehicles, emergency reserves, and other landers. Storing surplus oxygen improves the resource logic of hydrogen-oxygen propellant production because the plant produces more oxygen than Blue Moon would need under the central mixture assumption.
How long would a single refill take?
With 250 to 500 kilowatts of average delivered plant power, a central refill could take months after commissioning. With 1 megawatt of reliable delivered power and favorable ice, the energy portion could fit within one to four months. Low ice grade or low mining throughput could extend production beyond a year.
Would the first Blue Moon missions depend on lunar-made propellant?
Public NASA and Blue Origin information does not indicate that the first Blue Moon crewed missions depend on lunar-made propellant. Early Artemis lander operations focus on development, testing, demonstration, docking, crew transfer, and mission safety. Lunar-made propellant would require a separate surface industrial capability.
Why is liquid hydrogen harder than liquid oxygen?
Liquid hydrogen must be kept near 20 kelvin, far colder than liquid oxygen near 90 kelvin. Hydrogen’s low density and small molecules make storage and transfer difficult. A lunar plant must handle hydrogen liquefaction, tank conditioning, boiloff control, and leak management with limited maintenance access.
What is the biggest uncertainty in the estimate?
The biggest uncertainty is the quality of the ice deposit at the actual mining site. A high-grade, accessible deposit can reduce mining mass and timeline. A low-grade or difficult deposit can force much larger excavation volumes, more machine wear, higher extraction energy, and a slower refill campaign.
Appendix: Glossary of Key Terms
Artemis
Artemis is NASA’s program to return astronauts to the Moon, develop sustained lunar operations, and prepare systems for later Mars missions. It includes the Space Launch System, Orion, Gateway, commercial human landing systems, spacesuits, rovers, science payloads, and supporting infrastructure.
BE-7
BE-7 is Blue Origin’s liquid oxygen and liquid hydrogen engine for the Blue Moon lunar lander family. It is designed for high efficiency, deep throttling, and restart capability, which are needed for lunar descent, landing, ascent, and orbital maneuvering.
Blue Moon Mark 2
Blue Moon Mark 2 is Blue Origin’s crewed lunar lander design for NASA’s Artemis human landing system architecture. It is intended to carry astronauts between lunar orbit and the lunar surface, with NASA using a full-scale crew cabin mockup for training and mission simulations.
Electrolysis
Electrolysis is the process of using electricity to split water into hydrogen and oxygen. In a lunar propellant plant, electrolysis would turn mined and purified lunar water into the gas streams later cooled into liquid hydrogen and liquid oxygen.
Human Landing System
The Human Landing System is NASA’s term for the lander that transports astronauts from lunar orbit to the Moon’s surface and back. NASA has contracts with commercial providers to develop these systems for Artemis missions and sustained lunar surface access.
In-Situ Resource Utilization
In-situ resource utilization means using materials found at the destination instead of launching every supply from Earth. On the Moon, the most discussed examples include extracting water ice, producing oxygen, making propellant, and using regolith for construction materials.
Liquid Hydrogen
Liquid hydrogen is hydrogen cooled to an extremely low temperature so it becomes a liquid. It offers high rocket performance when burned with liquid oxygen, but it is difficult to store because it requires very cold conditions and careful boiloff management.
Liquid Oxygen
Liquid oxygen is oxygen cooled into liquid form for use as rocket oxidizer or life-support supply. It is easier to store than liquid hydrogen, though still cryogenic. In water-based lunar propellant production, oxygen is produced in surplus under many engine mixture assumptions.
Permanently Shadowed Region
A permanently shadowed region is a lunar area, usually near the poles, that receives no direct sunlight because of local terrain and the Moon’s low axial tilt. These regions can remain cold enough to preserve water ice for long periods.
Water Ice Grade
Water ice grade describes the share of water in icy regolith by mass. A 5% grade means 100 kilograms of material contains about 5 kilograms of water before recovery losses. Grade strongly affects mining volume, energy demand, and refill timeline.

