What is ISRU and how can it help humanity explore the solar system?

Let's look to the future. Not 100 years, not 50, but 10; to humanity's first permanent extraterrestrial settlement.

This lunar village will serve as the starting link in what will be a very long chain. A chain that supports a new economy and the final frontier, a chain that makes exploring the depths of the universe possible.

This is the first test. Can we learn to use the resources available to us in space? Can we free ourselves from our reliance on our mother planet?

For now, we remain like a dog on a leash, exiled to the corner of our yard.

We came to the Moon with structures that unfold like origami. With the help of robotics, we put together a new settlement like a game of high stakes Lincoln Logs. We have some of Earth's best and brightest scientists here, all working together to build the foundation for a fuel depot. We will harvest volatiles from permanently shadowed craters near the lunar poles. We will create fuel from the frozen molecules in the lunar regolith.

Before long, we will have a full fleet of reusable spacecraft that roam Earth orbit, refuelling the constellation of satellites that orbit the planet, and making a pretty penny doing so.

But this is not our goal. We dream of a deep and infinite space whose treasures lie far beyond our wildest dreams. The only way this goal is achieved is by creating a strong and flexible infrastructure for us to build upon, to iterate and to start again.

Here we stand on the crest of the first wave. I see the second on the horizon and I gape in awe. I can only imagine what the third will bring.

Human missions to the International Space Station (ISS) and the Moon have depended almost entirely on resources from Earth. All fuel, materials and food must be launched with the crew, and resupplied over time on long missions. Space habitats like the ISS are designed to reuse what they can, through air replenishment, CO2 processing and wastewater recycling. In 2023 the station's life support system showed it can recover about 98% of the water crews bring with them, the level NASA says long journeys need (NASA). As we set our sights on more distant destinations, our tether to Earth's resources must be reduced or cut entirely.

In-situ resource utilization (ISRU) means using the natural resources found at a mission's destination (NASA). ISRU technologies reduce dependence on supplies from Earth, and bring space habitats closer to closed systems that can sustain themselves.

Future settlements practising ISRU may also trade resources. Beyond a closed system, ISRU could take in-space manufacturing, refuelling and habitat construction much further. One way to sort the resources it could use is the way ecologists describe the non-living parts of an ecosystem: light, atmosphere, water and geology. This piece takes them in that order.

Light

Sunlight is a commonly used resource, both on Earth and in space. Solar panels are arrays of photovoltaic cells that convert light directly into electricity. Satellites and habitats in orbit rely on sunlight as their main source of energy.

On the Moon, sunlight has limits. Nights near the poles can last more than 14 Earth days, and some places never see the Sun at all. NASA plans to put a nuclear reactor on the Moon by the first quarter of fiscal year 2030. A reactor could be placed "in lunar regions where sunlight cannot reach" and keep running through the long nights (NASA). Solar power also gets weaker the further you travel from the Sun, so missions deeper into the solar system will need other sources of energy.

Atmosphere

The greatest potential for using a local atmosphere is on our planetary neighbour, Mars. The Martian atmosphere is 95.1% carbon dioxide, 2.59% nitrogen and 1.94% argon (NASA). It is only about one percent as thick as Earth's (NASA).

That carbon dioxide is a raw material. NASA's MOXIE instrument rode to Mars on the Perseverance rover and split Martian CO2 to make oxygen. Over 16 runs between 2021 and 2023 it made 122 grams in total. A future system would need to make oxygen in industrial quantities, for astronauts to breathe and for the rocket propellant to launch them home (NASA).

Water

Life as we know it cannot survive long without liquid water. Humans in space need a reliable supply of it. Water is also used to grow food, to make oxygen and in most industrial processes.

Mars has plenty of frozen water. Scientists say there is enough buried ice in one region of Mars to fill Lake Superior. Researchers have also tried pulling water out of the thin Martian air. WAVAR, a water vapour adsorption reactor developed at the University of Washington, was tested in Mars-type conditions, but there was no funding to take it beyond proof of concept (Cosmic Log).

On the Moon, the ice sits in the coldest places. Permanently shadowed areas exist at the bottom of deep craters near the south pole (NASA). In October 2009 NASA's LCROSS mission sent a spent rocket stage into Cabeus crater, near the south pole, and found water ice. The soil there was also rich in hydrogen gas, ammonia and methane, which could be used to make fuel (NASA).

Geology

Lunar settlements will need construction materials from lunar sources. Fine lunar soil can be sintered into compressed blocks for building. What the soil is made of depends on where you dig. In the dark plains, the maria, it is about 45% silica, 15% alumina, 14% iron oxide, 12% lime and 9% magnesia. In the highlands it has much more alumina, about 24%, and much less iron oxide, about 6% (NASA). Highlands are found all around the Moon, including the polar regions where future crews are headed (Maana Electric).

Lunar soil is also full of oxygen. It is 40 to 45% oxygen by weight, its most abundant element. But that oxygen is "bound up chemically as oxides in the form of minerals or glass", so it has to be extracted before anyone can use it (ESA). Oxygen from the soil could become rocket propellant. Oxygen makes up most of the propellant mass needed to lift off from the Moon and for chemical propulsion in space (NASA).

The lunar surface has also been bombarded with large quantities of helium-3 by the solar wind. Fusion studies have proposed that helium-3 could provide safer nuclear energy in a fusion reactor (ESA). That is still a proposal. The first buyers want helium-3 to cool quantum computers to very low temperatures. In May 2025 Interlune agreed to supply lunar helium-3 to the US Department of Energy and to Maybell Quantum Industries, starting by 2029 (Cosmic Log). In September 2025 Bluefors, which makes cryogenic cooling systems for quantum technology, agreed to buy up to ten thousand litres a year for delivery from 2028 to 2037 (Interlune).

Case study: Helios

A growing number of companies are developing ISRU technologies, and here we highlight one. Helios is an Israeli startup that set out to extract oxygen from lunar soil, to supply the oxidizer in rocket fuel.

Its key product was a molten regolith electrolysis reactor. The process melts lunar soil at 1,600 degrees Celsius and then uses electrolysis to make oxygen (Times of Israel). In our interview with Helios founder Jonathan Geifman, he described it as separating "the vast amount of oxygen from the various metals that constitute the regolith".

Helios has since changed direction. It now describes itself as a metal refining company, and says it has made iron from iron ore. Lunar oxygen is still listed among its products (Helios).

How would an abundant supply of fuel on the Moon help humanity become multi-planetary?

Oxygen is the heaviest part of a rocket's propellant. In a 2025 paper, engineers from Sierra Space and NASA wrote that oxygen "typically constitutes over 70% of the mass of the vehicle" (NASA). Oxygen made on the Moon would not have to be launched out of Earth's gravity. It could fuel spacecraft on the Moon, in lunar orbit and beyond.

ISRU would decrease the mass of fuel and materials that need to be shipped to orbit, the lunar surface or Mars. That leaves more of each launch for other uses, and should reduce the cost of space travel overall. Fully operational ISRU would also make a space economy less exposed to supply chain problems on Earth. A shortage of a key material on Earth would not affect a lunar habitat that makes its own.

The science and its applications are still at the research and development stage. MOXIE made oxygen on Mars. On the Moon, the first attempt to drill for ice ended early. NASA's PRIME-1 drill reached the south pole region on Intuitive Machines' IM-2 lander in March 2025, but the lander came down on its side in a crater and the mission was over in about 13 hours. The drill went through its full range of motion without reaching the soil (SpacePolicyOnline). Tests on Earth use simulated soil. In February 2026 a NASA team used concentrated sunlight to extract oxygen from simulated lunar soil (NASA).

If ISRU can make space cheaper, commercial players will put their resources into it, and they already are. Governments are still the main customers, but Interlune's helium-3 deals show private buyers are starting to sign up too. Living off-planet for long will depend on using what is already there.

Where things stand in October 2026

NASA has made the Moon's south pole the centre of its plans. Artemis II flew four astronauts around the Moon in April 2026 (NASA). The first landing has moved to Artemis IV, planned for 2028. Artemis III will test the lander in Earth orbit in 2027 (The Register). In March 2026 NASA announced a plan for a permanent human presence at the lunar south pole by 2036. The region is thought to hold ice and other valuable materials in its shadowed craters (Scientific American).

Robotic missions come first. NASA's VIPER rover was cancelled in July 2024 and revived in September 2025 (Astronomy). It is scheduled to reach the south pole in late 2027 on Blue Origin's second Blue Moon lander, to search for and map water ice (NASA). The Griffin-1 lander is due to launch on a Falcon Heavy in the fourth quarter of 2026 (Spaceflight Now). It carries METAL, a NASA instrument developed with Interlune to measure helium-3 in the soil (NASA). It also carries Astrolab's FLIP rover. The rover carries an Interlune camera whose images will be used to estimate how much helium-3 the soil holds (Astrolab).

Not everything has worked. ispace's second lunar lander, Resilience, crashed in June 2025. A preliminary analysis found that its laser system for measuring altitude did not work as planned, and the lander came down too fast (NBC News). It was carrying a water electrolyzer from Takasago Thermal Engineering (ispace). ispace's next landing is scheduled for 2028 (ispace). In July 2026 ESA hired its European arm to deliver MAGPIE, Europe's first rover to look for ice at the lunar poles, due to fly in 2029 (ispace). In August 2026 China postponed Chang'e-7, a mission to explore permanently shadowed craters at the south pole. The space agency said it could not launch during the planned window this year (AP via ClickOnDetroit).

Interlune is moving ahead on helium-3. In May 2026 NASA hired it to build a payload, ready for launch in 2028, that would collect lunar soil and extract the gases the solar wind left in it (Interlune). In July 2026 it said its Cold Capture technology had produced 99% pure helium-3 from Grade A helium on Earth (Interlune).

Other companies in our directory work on the pieces around ISRU. In October 2025 Maana Electric and its partners collected, moved and processed lunar soil simulants in a field test at ESA's LUNA facility in Cologne (Maana Electric). NASA has selected Astrolab and Lunar Outpost to provide the first fleet of lunar terrain vehicles for Artemis astronauts (NASA). Further out, AstroForge prospects asteroids for metals, and Karman+ plans to process asteroid material in space and use it to refuel spacecraft in orbit (Payload).

The companies in this piece

Updated 9 October 2026. We corrected and updated this post. The Moon's ice sits in permanently shadowed craters near the poles, not on a "dark side". The figures for Mars's atmosphere, lunar soil and its oxygen now follow NASA and ESA, and soil from the lunar highlands differs from the plains. Lunar oxygen helps because it would not have to be launched from Earth, which the original described wrongly. We took out two unsourced claims, about heat from Io and design limits, and a helium-3 diagram we could not credit properly. We also updated Helios, added helium-3's first customers, and added where things stand in October 2026.