The research team, led by James E. Johnson together with George F. Sowers and Angel Abbud-Madrid, built its assessment around two of the most mature habitation elements planned for NASA’s Moon Base: the Japan Aerospace Exploration Agency’s Pressurized Rover and the Italian Space Agency’s Multi-Purpose Habitation module. Both concepts have advanced into preliminary design, making them credible candidates for the earliest human outposts on the surface. Because these elements face strict limits on mass, power, thermal control, and internal volume, they are expected to fly with open-loop, non-regenerative life support systems rather than the water-recycling and oxygen-recovery hardware used on the International Space Station. That design choice means every breath and every liter of hygiene water must be shipped from Earth, at least until local production comes online.

To quantify what that dependence costs, the researchers modeled consumable demand across minimum, baseline, and maximum mission architectures, varying crew size from two to four, mission duration, and the frequency of extravehicular activities. Demand drivers included metabolic oxygen consumption, drinking and food preparation water, hygiene and flush water, cabin leakage, and, critically, the gas lost every time an airlock is cycled for a spacewalk. Using the ideal gas law and Dalton’s law for a two-gas atmosphere, the team calculated the mass of oxygen and nitrogen expelled during each repressurization event, a contribution that earlier studies had overlooked. The result was a baseline oxygen demand 175 percent higher than prior estimates that counted only metabolic use, while baseline water demand came in 45 percent lower than earlier work because it excluded crew showers, a capability abandoned in human spaceflight since the Skylab era.

The analysis then confronted a problem that dominates the economics of shipping anything to the Moon: containment and packaging. Drawing on International Space Station logistics, the study accounted for high-pressure gas tanks for oxygen, contingency water carriers for water, and the soft-sided, foam-lined cargo transfer bags in which both are packed for transit. These containers add enormous overhead. For oxygen, tanks and packaging inflate delivered mass by roughly 229 percent above the raw consumable demand, compared with about 21 percent for water. Because carriers are assumed to be filled to capacity regardless of exact demand, excess consumables ride along as operational reserves, further increasing the mass that must be launched, landed, and handled on the surface. This packaging penalty, the authors emphasize, is the primary reason Earth-based resupply is so inefficient for early habitation.

With demand and delivered mass established, the team estimated lifecycle costs using three independent parametric approaches: NASA’s Advanced Missions Cost Model, a regression-based tool built from more than 260 historical spaceflight projects; NASA’s Project Cost Estimating Capability, which applies cost-estimating relationships within a system-level work breakdown structure; and a simplified industry-based method using heuristic cost factors of $63,000 per kilogram for development and $25,200 per kilogram for flight unit production, adjusted to fiscal year 2026 dollars and coupled with a 90 percent learning curve. Delivery costs were fixed at $100,000 per kilogram, a figure consistent with SpaceX’s most recent projection for lunar surface delivery and far below the roughly $1 million per kilogram implied by current Commercial Lunar Payload Services contracts. Operations costs were applied as a wrap factor of 5.5 percent of development and production costs per year, a deliberately conservative lower bound drawn from published ranges.

Normalizing the median lifecycle cost by delivered consumable mass produced the study’s headline numbers: approximately $286,000 per kilogram for oxygen, roughly $121,000 per kilogram for water, and about $185,000 per kilogram for a combined oxygen-and-water resupply architecture. Oxygen’s premium, more than twice that of water despite lower demand by mass, reflects the brutal efficiency of its packaging, since high-pressure tanks weigh far more relative to the gas they carry than water bags do. The authors note that these figures represent the maximum price a government habitation customer might rationally pay for lunar-derived consumables before preferring Earth resupply, and they caution that the estimates may even be conservative because supporting infrastructure such as pressurized transport systems was not included.

Against these values, the researchers compared three pilot-scale ISRU concepts: a water ice plant that electrolyzes extracted ice, a carbothermal reduction system that processes regolith, and a molten salt electrolysis system that extracts oxygen directly from regolith. Each was sized for roughly 1,000 kilograms of oxygen production per year, with subsystem-level adjustments for habitation service, such as removing liquefaction hardware, adding high-pressure compression for spacesuit recharge, and including a water processor assembly when potable water must be delivered. The comparison revealed a clear economy-of-scale effect: in all but the highest demand scenarios, Earth resupply remains cheaper, but as mission cadence and crew size grow, ISRU production costs fall per kilogram and become competitive, particularly for oxygen. A single water ice or carbothermal pilot plant can meet per-mission oxygen and water demand in nearly all modeled cases, and a 50 percent reduction in assumed recurring spares mass would improve ISRU competitiveness by up to $65 million in lifecycle cost.

The strategic implications extend well beyond a single contract. Previous propulsion-focused ISRU business cases assumed lunar oxygen could command only $500 to $35,000 per kilogram when sold as oxidizer to vehicles operating in cislunar space. A habitation customer standing on the lunar surface, by contrast, is expected to pay an order of magnitude or more than those figures because of the crushing logistical penalties of Earth resupply. The authors argue that this near-term, small-volume market could serve as the catalyst that funds the first ISRU demonstrations, builds operational experience, and reduces technical risk in advance of the much larger propellant market, which some analyses project could approach a $63 billion valuation by 2040. NASA’s own Moon Base planning anticipates ISRU experimentation through 2032 with scalable production beginning around 2033, a timeline that aligns closely with the emergence of the habitation demand modeled in this study.

The study also acknowledges important uncertainties. The Advanced Missions Cost Model, rooted in government programs predating 1999, produced estimates 23 to 48 percent higher than the other two approaches, reflecting the cost reductions achieved through fixed-price contracting, reusability, and additive manufacturing in the commercial era. The pace at which early missions evolve toward larger crews, aggregated habitats, and regenerative life support systems remains uncertain, dependent on technology development, mission success, funding, and geopolitical factors. If future habitats add water recovery approaching 98 percent efficiency, the demand picture would shift substantially, though the authors show their methodology can be extended to those architectures. Comparable analyses could also be applied to China’s International Lunar Research Station once detailed designs become available.

What emerges is a compelling and surprisingly practical vision for the first lunar economy. Rather than waiting for a sprawling propellant infrastructure that may take decades to materialize, entrepreneurs and space agencies now have a quantified, near-term target: small habitats on the Moon that will pay hundreds of thousands of dollars per kilogram for oxygen and water if someone can produce them locally. The extrinsic benefits, including risk reduction, technology maturation, and first-to-market advantage, may further attract venture capital and public-private partnerships. If pilot-scale ISRU systems can deliver consumables at a lower lifecycle cost per kilogram than Earth resupply, the path to a sustained lunar economy may begin not with rockets refueling at depots, but with a modest plant quietly making breathable air and clean water for the crews who live there.

Subject of Research: Lifecycle cost valuation of lunar-derived oxygen and water for early Moon Base habitation as an initial market for in-situ resource utilization

Article Title: The value of lunar-derived oxygen and water for an early habitation customer

Article References: Johnson, J. E., Sowers, G. F., & Abbud-Madrid, A. (2026). The value of lunar-derived oxygen and water for an early habitation customer. Space and Planetary Resources, 2(1), Article 12. https://doi.org/10.1007/s44461-026-00017-8

Image Credits: AI Generated

DOI: 10.1007/s44461-026-00017-8

Keywords: in-situ resource utilization, lunar habitation, Moon Base, oxygen and water demand, lifecycle cost analysis, space logistics, cislunar economy, lunar ISRU, NASA Artemis, pilot plant, space resources, lunar economy

Structural health monitoring is already a mature discipline on Earth. Engineers define damage as any change to a structure’s material properties, geometry, or internal connectivity that degrades performance, even if the system still functions. An SHM system typically combines four elements: sensor instrumentation measuring strain, displacement, vibration, temperature, and humidity; a data acquisition system that digitizes the signals; analysis software using signal processing, statistics, and machine learning to extract meaning; and reporting tools that let engineers assess risk and schedule maintenance. On the Moon, the authors argue, these components become an early-warning platform that lets astronauts and ground controllers shut down compromised zones, isolate failures, and act before a microfracture in a pressurized habitat escalates into catastrophe.

The lunar environment makes the case urgent. Surface temperatures swing from roughly 387 kelvin in daylight to about 95 kelvin at night, while areas near permanently shadowed regions at the poles can plunge below 90 kelvin. Moonquakes and meteoroid impacts generate vibrations, and launch or landing operations impose intense acoustic and mechanical loads. Lunar dust, which is abrasive and electrostatically adhesive, coats structures and disrupts thermal balance, while volatile chemicals and space-weathering ions can alter regolith at the microscopic scale, changing porosity and diffusivity beneath foundations. Because the interactions between the dusty plasma environment, thermal cycling, and multi-material structures remain poorly characterized, the authors stress that real-time monitoring is the only practical way to track damage as it develops.

The review draws heavily on terrestrial analogs. Fiber Bragg grating sensors have protected historic churches and monitored the Geumdang Bridge in South Korea, where abnormal strain patterns prompted reinforcement before degradation spread. Acoustic emission arrays in South African gold mines and European underground laboratories detect millimeter-scale microseismic activity in rock up to 200 meters away, a capability directly relevant to lunar tunnels and subsurface habitats. Cryogenic storage tanks validated piezoelectric sensors under extreme cold, and Alaska’s Trans-Alaska Pipeline demonstrates low-power, autonomous monitoring under seasonal thermal extremes and remote, power-limited conditions that closely resemble the lunar poles. Even a cautionary tale appears: the 2004 collapse of Singapore’s Nicoll Highway, attributed partly to undetected soil movement during excavation, underscores what happens without adequate monitoring.

Space heritage matters too. The first documented SHM in a space mission came in 1997, when a McDonnell Douglas Delta rocket carried fiber Bragg grating strain sensors. NASA’s shuttle modal inspection system hunted for fatigue damage in fuselage panels, and piezoelectric wafer active sensors proved effective at interrogating bolted joints in thermal protection panels, where loosening from micrometeoroid strikes and thermal cycling could prove fatal. The International Space Station, assembled from 32 modules over 11 years, is the closest existing model of space architecture, and its modularity offers lunar builders flexibility. Yet the authors caution that the ISS operates in microgravity with very different materials, so its sensor solutions cannot simply be copied for the lunar surface.

Sensor choice will depend on facility type, power availability, and the specific failure modes anticipated. Fiber-optic sensors carry a low weight penalty, resist electromagnetic interference, tolerate wide thermal ranges, and allow many sensing points along a single cable. Piezoelectric wafer sensors offer compact, low-power ultrasonic interrogation for locating damage in composites and thin-walled structures. Wireless networks, supporting up to 15 channels and roughly eight kilometers of outdoor range, eliminate cabling that once tripped Apollo astronauts and overheated under insulating dust. Pressure sensors guard habitat integrity, vibration sensors distinguish moonquakes from machinery faults, and dust sensors track filter-clogging accumulation. Battery technology matters as much as sensing: lithium-polymer packs can run devices for up to a year, and paired with high-capacity batteries or solar panels, some systems have demonstrated potential ten-year lifetimes without human intervention.

The heart of the paper is a set of case scenarios for coordinated, site-wide monitoring. In one, a communication relay tower stands several hundred meters from a reusable launch pad; strain gauges, tilt sensors, and accelerometers on both structures let operators separate genuine tower degradation from launch-induced shaking, eventually defining safe standoff distances. In another, automated mining machinery operates within 100 meters of a pressurized underground tunnel; fiber-optic strain sensors and tiltmeters fused with mining-platform load and thermal data reveal ground settlement and fatigue before tunnels are crossed. A third places a thermal research station near a permanently shadowed region beside a waste-management facility, where cross-facility sensors guard against localized warming that could compromise ultra-cold experiments or materials.

The authors also frame SHM within a phased decision process for lunar development: site selection, master planning, site work and infrastructure installation, building construction, and long-term maintenance and rehabilitation. They recommend embedding SHM planning into the master planning stage, choosing sensors and their numbers early, especially for complex sites with multiple interacting operations. A preliminary decision framework matches sensor types to facility classes, from pressure-critical laboratories and habitats to regolith-covered tunnels, chemical processing areas, mineral extraction units, rover storage bays, and exposed solar arrays vulnerable to dust, hot spots, and micrometeoroid strikes. Damage prognosis then quantifies fatigue cycles before criticality, distinguishing gradual wear such as corrosion and dust accumulation from predictable discrete events like launch loading and unpredictable shocks like meteoroid impacts.

Deployment will evolve with mission duration. Short-term missions of up to 30 days call for rapidly deployable, surface-mounted MEMS accelerometers, strain gauges, and thermocouples with edge computing for autonomous event detection. Intermediate missions of one to twelve months shift to embedded fiber-optic and piezoelectric arrays with redundancy and hybrid power to survive lunar night. Long-term settlements will require hierarchical networks fusing distributed nodes with centralized processing, artificial intelligence for trend analysis and predictive maintenance, and modular sensor packages designed for robotic or spacewalk replacement. The trade-off between external sensors, which are easy to retrofit but vulnerable to dust and radiation, and embedded sensors, which are sensitive and protected but nearly impossible to replace, will shape construction methods themselves.

Ultimately, the authors argue, SHM should be treated as a cross-cutting infrastructure layer, as fundamental to a lunar base as power, communications, and mobility. Beyond safety, it enables condition-based maintenance that conserves scarce resources, extends facility lifespans, and lowers total mission cost, while the data it gathers feeds back into better materials, refined 3D-printing and sintering techniques, and smarter designs for the next generation of habitats. As humanity prepares for a sustained presence on the Moon, the quiet hum of a sensor network may prove to be the difference between a thriving settlement and a structural failure 380,000 kilometers from the nearest repair crew.

Subject of Research: Structural health monitoring sensor systems for future lunar base architectures

Article Title: Case scenarios for structural health monitoring for future lunar architectures

Article References: Ahrens, C., & Hollander, J. (2025). Case scenarios for structural health monitoring for future lunar architectures. Space and Planetary Resources, 1(1), Article 3. https://doi.org/10.1007/s44461-025-00003-6

Image Credits: AI Generated

DOI: 10.1007/s44461-025-00003-6

Keywords: lunar architecture, structural health monitoring, sensors, moon base, damage prognosis, fiber-optic sensors, piezoelectric sensors, lunar dust, in-situ resource utilization, space habitats, machine learning, lunar infrastructure

The material at the heart of this work is lunar regolith, the loose, fragmented layer of rock, mineral grains, and glassy particles that blankets the lunar surface to depths of several meters. Regolith is the product of billions of years of meteorite impacts that pulverized the lunar crust, and its composition varies across the Moon but generally includes silicate minerals, oxides of iron, titanium, calcium, and aluminum, and a significant fraction of agglutinates, which are irregular glassy particles welded together by micrometeorite impacts. Because this material is already sitting on the lunar surface in essentially unlimited quantities, it represents the single most obvious feedstock for any serious attempt at in-situ resource utilization, the strategy of living off the land beyond Earth.

Additive manufacturing, more commonly known as 3D printing, offers a natural fit for this challenge. Unlike conventional construction, which relies on large machinery, formwork, and a skilled workforce, additive manufacturing builds structures layer by layer from a digital design, using only the material that is actually needed. On the Moon, where every machine must be shipped from Earth and operated in a vacuum, under extreme temperature swings, and amid abrasive dust, the appeal of a compact, automated, digitally controlled fabrication system is hard to overstate. A single printer, paired with a regolith harvesting and processing system, could in principle fabricate walls, foundations, radiation shields, landing pads, and infrastructure components on demand, adapting each design to local terrain and mission requirements without waiting for resupply missions.

What distinguishes the new study is its emphasis on reconfigurability. Most visions of printed lunar habitats assume a one-way process: a structure is designed, printed, and fixed in place forever. But mission planners increasingly recognize that lunar bases, like the missions that precede them, will need to evolve. Equipment will be replaced, modules will be repurposed, and habitats will need to expand or contract as crew rotations and scientific priorities change. Building blocks that can be printed, assembled, disassembled, and reassembled into new configurations would give lunar architects a flexibility that monolithic printed structures cannot provide. Instead of demolishing a wall to build a new room, crews could simply take the wall apart and reprint or reposition its elements elsewhere.

Achieving this vision requires solving a chain of interlocking technical problems, and the researchers address them across the full workflow. The first step is feedstock preparation. Raw lunar regolith, whether actual Apollo-era samples or, more commonly in laboratory research, lunar regolith simulants that replicate the mineralogy and particle size distribution of the real material, must be sieved, sorted, and in some cases processed into a form suitable for printing. The sharp, irregular, and glassy nature of regolith particles makes them abrasive and difficult to flow uniformly, so particle engineering plays a crucial role in producing a feedstock that a printer can handle reliably.

The second step is the printing process itself, and here the study examines how regolith-based materials behave when deposited layer by layer. A central tension in lunar construction chemistry is the binder problem. On Earth, concrete gains its strength from Portland cement, whose production requires water and generates carbon dioxide through the calcination of limestone. Neither the water nor the emissions are acceptable on the Moon, where water is a precious resource and there is no atmosphere to pollute. Alternatives under investigation across the field include geopolymer chemistry, in which alkaline solutions activate the aluminosilicate minerals in regolith to form cement-like binders; sintering, in which concentrated heat from lasers, microwaves, or focused sunlight fuses regolith particles into solid masses without any binder at all; and small quantities of imported bonding agents, such as polymers, used as economically as possible.

Each route involves trade-offs that the researchers weigh in detail. Sintering produces genuinely binder-free structures, a major advantage for long-term self-sufficiency, but the vacuum environment complicates heat transfer and can trap gases released from the regolith, causing porosity and cracking. Thermal expansion mismatches between layers and the extreme thermal cycling between lunar day and night, where surface temperatures can swing by more than two hundred degrees Celsius, add further stresses. Geopolymers and chemical binders can deliver strong, dense components at lower processing temperatures, but they introduce dependence on reactants that must either be sourced locally or transported from Earth. The study’s framework for reconfigurable blocks is designed to accommodate this uncertainty: because the blocks are modular, a printing process can be refined or even replaced over time without abandoning the structures already built from earlier batches.

Mechanical performance is, of course, the bottom line for any structural material, and the reported work includes evaluation of the printed blocks under conditions relevant to lunar service. Compressive strength is the primary metric, since lunar habitats will mostly experience compressive loads from overlying regolith shielding piled on top of habitats to protect crews from galactic cosmic rays and solar particle events. Several meters of regolith cover are typically proposed for radiation protection, which means the underlying structure must bear substantial static loads in one-sixth of Earth’s gravity. The blocks must also tolerate internal pressurization, because habitats will hold breathable atmosphere at pressures that push outward on the walls, creating tensile stresses that brittle, sintered regolith handles poorly. Strategies to address this include placing habitat pressure vessels inside regolith-block shells, reinforcing blocks with fibers or mesh, and designing interlocking geometries that distribute loads across many contact surfaces rather than relying on mortar joints.

The interlocking geometry is where the reconfigurable concept becomes tangible. Rather than printing large monolithic panels, the researchers envision blocks with engineered shapes, analogous to LEGO bricks or precision masonry units, that can be stacked into curved walls, domes, and vaults and later separated without destructive force. Digital design tools allow each block’s geometry to be optimized for its position in a structure, embedding channels for cables and pipes, sockets for mounting hardware, or keying features that align with robotic grippers. This last point matters because much of the assembly on the Moon will likely be performed by robots rather than astronauts. Robotic arms placing regolith blocks in a vacuum environment avoid the hazards of EVA, and modularity suits robotic manipulation far better than amorphous printed masses, since discrete units with well-defined geometry can be grasped, positioned, and verified with existing machine-vision techniques.

Looking toward actual missions, the researchers situate their work within the broader context of NASA’s Artemis program and international plans for a sustained lunar presence, including the proposed Moon Village concept championed by the European Space Agency. The surface of the Moon is expected to host multiple cooperating installations in the coming decades, from the Gateway-linked Artemis Base Camp at the lunar south pole to landing infrastructure, power plants, telescopes, and pilot plants for extracting oxygen and metals from regolith. All of these will need construction materials, radiation shielding, thermal management, and foundations, and all of them will benefit from a standardized, printable, reconfigurable building system. The authors position their building blocks not as a finished habitat but as a scalable construction primitive, a verified unit of lunar architecture around which future designs, standards, and robotic systems can converge.

Significant engineering hurdles remain before regolith blocks are stacked on the lunar surface. Testing with genuine lunar samples is rare and limited by the tiny quantities of Apollo material available, so validation ultimately depends on simulants whose fidelity to the real thing is imperfect and whose behavior under vacuum, radiation, and thermal cycling differs in ways that are still being characterized. Printing at useful scale in vacuum, with lunar gravity and without Earthlike supply chains, has yet to be demonstrated in an operational setting, although parabolic flights and vacuum-chamber experiments continue to close the gap. The new study contributes a coherent pathway through this landscape: a demonstration that regolith can be additively manufactured into discrete, mechanically sound, reconfigurable blocks, and a design philosophy in which habitats grow and change with the missions they serve. If the approach matures as hoped, the first permanent structures on the Moon may not be transported there at all, but printed in place from the ground beneath future astronauts’ boots, one reconfigurable block at a time.

Subject of Research: Additive manufacturing of lunar regolith into reconfigurable building blocks for lunar habitation.

Article Title: Additive manufacturing of lunar regolith for reconfigurable building blocks toward lunar habitation

Article References: McCallum, C., Liang, Y., Tushar, N., Xu, B., Zhao, B., Zeng, H., & Shou, W. (2026). Additive manufacturing of lunar regolith for reconfigurable building blocks toward lunar habitation. npj Advanced Manufacturing. https://doi.org/10.1038/s44334-026-00111-x

Image Credits: AI Generated

DOI: 10.1038/s44334-026-00111-x

Keywords: lunar regolith, additive manufacturing, 3D printing, lunar habitation, in-situ resource utilization, Moon base, space construction, sintering, geopolymer binders, reconfigurable building blocks, Artemis program, radiation shielding

The raw material in question is lunar regolith, the powdery, shattered blanket of rock that covers the Moon’s surface. Forged by billions of years of micrometeoroid impacts and space weathering, regolith is a complex mixture of silicate minerals and metal oxides, including silicon dioxide, aluminum oxide, calcium oxide, iron oxide, titanium dioxide, and magnesium oxide. Locked inside these oxides are exactly the elements a lunar base needs most: oxygen for breathing and rocket propellant, and metals for construction, fabrication, and power infrastructure. But the chemistry is unforgiving. These elements are bound tightly in mineral matrices, and liberating them demands either extreme heat or aggressive electrochemical processing. The Moon’s own environment compounds the challenge, offering high vacuum, reduced gravity, abrasive dust, and a fourteen-day night that strains any power system.

Where you dig matters enormously. The dark volcanic plains of the lunar maria are enriched in iron, titanium, and magnesium, making them attractive for processes that target those oxides. The brighter highlands, by contrast, are dominated by anorthosite, a rock built largely from the aluminum- and calcium-rich feldspar anorthite. Because aluminum is prized for structural applications, electrical conductors, and recyclability, the research team focused its comparison on anorthite as a representative highlands feedstock, giving all three extraction methods an identical starting material so that differences in performance could be attributed to the processes themselves rather than to geochemistry.

The first contender is carbothermal reduction, or CTR, a high-temperature chemical process that uses carbon, typically as graphite or methane, to strip oxygen from metal oxides. In theory, certain oxides can be reduced near 850 degrees Celsius, but practical operation generally demands temperatures above 1500 degrees Celsius, where partial melting of the feedstock improves reactant mobility and conversion. CTR splits into two regimes: a solid-gas pathway below roughly 1200 degrees Celsius with slower kinetics and incomplete conversion, and a molten-phase pathway above about 1400 degrees Celsius that delivers more extensive reduction. The process is especially attractive for silicon and iron production from iron-rich mare regolith, and it pairs naturally with concentrated solar thermal heating, reducing the demand for electrical power. Its Achilles heel is carbon. The Moon has essentially none, so large-scale CTR would require either constant resupply from Earth or in-situ carbon production, both of which add logistical complexity. Carbon can also react with silicon products to form silicon carbide, threatening both product purity and reactor longevity. Prior studies of carbothermal reduction on lunar regolith report oxygen yields of only 2.4 to 7.9 percent of the feedstock mass, corresponding to just 5.4 to 17.5 percent of the total available oxygen.

The second contender is molten salt electrolysis, or MSE, an electrochemical approach in which metal oxides are reduced inside a molten salt electrolyte, most commonly calcium chloride. Operating at a comparatively moderate 800 to 1000 degrees Celsius, MSE follows the logic of the FFC Cambridge process, in which oxygen ions are pulled directly out of solid oxides. Metal cations migrate to the cathode and deposit as metal, while oxygen gas evolves at the anode. The study highlights a lunar-adapted configuration called LISAP-MSE, for Lunar In-Situ Aluminum Production via Molten Salt Electrolysis, which first leaches anorthite with hydrochloric acid, thermally decomposes the resulting aluminum chloride hydrate into alumina, and then electrolytically reduces that alumina to metallic aluminum and oxygen. The calcium chloride electrolyte may even be derivable from processed regolith itself. MSE’s great strength is selectivity: because only electrochemically active oxides are reduced, beneficiation steps such as acid leaching, flotation, or magnetic separation can be used upstream to boost feedstock quality and deliver high-purity metals. The trade-offs are system complexity, slower oxygen production rates, and the need to manage electrode degradation and electrolyte stability over long durations in a dusty environment.

The third contender, molten regolith electrolysis, or MRE, is the most direct of all. Rather than beneficiating feedstock or adding an electrolyte, MRE simply melts raw regolith above roughly 1300 degrees Celsius, where the silicate melt becomes ionically conductive, and applies an electric potential across it. Metal cations such as iron, titanium, silicon, aluminum, and calcium are reduced at the cathode into mixed alloys, while oxygen streams off at the anode. This brute-force simplicity makes MRE appealing for early-stage missions where mass efficiency matters and oxygen is the priority, and it can in principle process the complete range of oxides present in lunar soil. But the same extreme temperatures impose severe engineering burdens: refractory containment materials must resist corrosion by molten silicate, electrodes must survive repeated thermal cycling and oxidation, and continuous power delivery is essential to keep the melt conductive and the reaction progressing.

To compare the three fairly, the researchers built a rigorous thermodynamic framework using temperature-adjusted enthalpy, entropy, and Gibbs free energy, drawing on NIST thermochemical data supplemented by specialized datasets for anorthite, steam, and graphitic carbon, whose high-temperature heat capacities required alternative formulations. Each process was modeled as complete conversion of anorthite to its products, establishing an upper-bound theoretical limit for aluminum and oxygen extraction. The results were normalized per kilogram of aluminum and oxygen produced, allowing a direct, first-principles comparison of inherent energy requirements at each process’s realistic operating temperature.

The verdict was unambiguous. All three processes are endothermic and require external energy input, but LISAP-MSE posted the lowest enthalpy and Gibbs free energy values of the three, making it the most energetically efficient route to simultaneous aluminum and oxygen production under the study’s assumptions. CTR landed in the middle, with higher thermal demands compounded by its consumable carbon feedstock and logistical overhead. MRE fared worst on energy, displaying the highest enthalpy and Gibbs free energy values, a reflection of the enormous cost of maintaining a fully molten silicate phase and driving electrochemistry within it. Notably, none of the reactions were thermodynamically spontaneous at the evaluated temperatures and one bar pressure, confirming that all of these processes will depend on sustained external heating and electrical power. The authors caution, however, that the ranking is framework-dependent: if bulk oxygen were the dominant goal, or if iron- and titanium-rich mare regolith were the feedstock, CTR and MRE could climb back up the rankings, and hydrogen reduction, excluded from this comparison, might also become attractive.

Perhaps the study’s most consequential insight is that the three technologies are not competitors at all, but natural partners. CTR and MRE share similar high-temperature operating conditions and produce high-throughput oxygen along with mixed-metal alloys of limited purity, while MSE operates cooler and refines selectively. The researchers sketch integrated architectures in which calcium aluminate byproducts from CTR and MRE are fed into an MSE module as beneficiated feedstock, yielding a high-purity calcium-aluminum alloy that then acts as a metallothermic reducing agent to liberate additional silicon and iron from leftover oxides, with the reformed calcium aluminates looping back to regenerate the alloy and release more oxygen. A second synergy uses the MSE-produced alloy as a pre-treatment for MRE, preferentially displacing iron, magnesium, and titanium from the melt to leave a silicon-enriched mixture that electrolyzes more cleanly. On the infrastructure side, the authors map out a phased concept of operations: deploy modular reactors with imported spares, pilot CTR or MRE for early oxygen, integrate autonomous excavation and hauling at rates of at least 2.54 kilograms of regolith per hour for a modest 10,000-kilogram-per-year oxygen plant, then scale up MSE modules as aluminum demand for construction and shielding grows. Power infrastructure must reach tens to hundreds of kilowatts, with nuclear fission surface power flagged as the most practical near-term backbone, supplemented by solar arrays during the lunar day. Autonomy will be essential given communication delays, and dust mitigation, through sealed pneumatic transfer, dust-tolerant seals, and electrodynamic dust shields, will decide whether reactors survive the abrasive lunar surface. The team’s recommendations for future research center on three fronts: materials durability in molten environments, continuous autonomous feedstock handling, and system-level integration of thermal, electrical, and mass flows. The message is clear: no single process can carry a lunar economy alone, and the first self-sustaining Moon base will be built on the clever coupling of all of them.

Subject of Research: Comparative thermodynamic analysis of carbothermal reduction, molten salt electrolysis, and molten regolith electrolysis for extracting aluminum and oxygen from lunar regolith

Article Title: A review and analysis of extraction methods for aluminum and oxygen from lunar regolith

Article References: Ortega, J. N., Rupp, B., & Han, F. D. (2026). A review and analysis of extraction methods for aluminum and oxygen from lunar regolith. Space and Planetary Resources, 2(1), Article 5. https://doi.org/10.1007/s44461-026-00010-1

Image Credits: AI Generated

DOI: 10.1007/s44461-026-00010-1

Keywords: lunar regolith, in-situ resource utilization, aluminum extraction, oxygen production, carbothermal reduction, molten salt electrolysis, molten regolith electrolysis, anorthite, lunar ISRU, space resources, thermodynamics, Moon base