JAXA MMX Mission: Why Japan Is Bringing Dust From Mars’ Moon Back to Earth

Twenty-two kilometers. That is roughly the average diameter of Phobos, JAXA MMX mission official site .the larger and inner of Mars's two small moons  a lumpy, cratered rock you could walk around in an afternoon, if walking on Phobos meant anything close to walking on Earth (it does not, and we will get to why). In October 2026, Japan's space agency JAXA plans to launch a spacecraft weighing about 4.5 tonnes toward this unassuming object, settle it into orbit around Mars, land on Phobos, scrape and core a few grams of its surface into a sealed container, and then fly that container across several hundred million kilometers of interplanetary space back to Earth. The mission is called MMX, for Martian Moons eXploration, and its entire premise rests on a stubborn question that four decades of orbital reconnaissance around Mars have failed to close out: where did this moon actually come from?

It sounds like it should be simple. Phobos is right there, photographed by half a dozen Mars orbiters, mapped down to its individual boulders. But composition and origin are different problems, and remote instruments  cameras, spectrometers, radar  can only infer so much about a surface from a few hundred kilometers away. MMX exists because scientists have concluded that the only way to settle this particular argument is to bring a piece of Phobos home and put it under instruments no spacecraft could ever carry.

MMX is led by JAXA,with the Japan Aerospace Exploration Agency serving as prime architect of the spacecraft, the mission design, and the sampling hardware for the JAXA MMX mission . It is not a solo effort. NASA contributes a gamma-ray and neutron spectrometer built at Johns Hopkins University's Applied Physics Laboratory, along with navigation and science collaboration. ESA supplies part of the deep-space communications system, including a transponder and power amplifier, and will lend its Estrack ground station network for tracking. CNES and DLR  the French and German space agencies  jointly built a small rover that will ride along and become the first machine to ever touch the surface of a Martian moon. The mission's main targets are Phobos and Deimos, but they are not treated equally: Phobos gets the landings, the drilling, and the sample collection, while Deimos will be studied from a distance during flybys on the way out of the system. That distinction matters. This is not another Mars orbiter cataloguing dust storms and polar caps. It is a mission built around touching one specific, tiny, strange object and carrying a piece of it home.

The mystery driving all of this splits into two competing stories. One holds that Phobos and Deimos are asteroids, wanderers from farther out in the solar system that strayed too close to Mars and were caught by its gravity, becoming permanent captives. The other holds that both moons are debris  the shattered remains of a large impactor, or perhaps a chunk of Mars's own crust, blasted into orbit by a catastrophic collision early in the planet's history, later reassembling into two small satellites the way Earth's Moon is thought to have formed from impact debris on a much larger scale. Neither explanation has been proven, and each has evidence pulling in its favor.

The case for capture leans heavily on what Phobos looks like from orbit: dark, reddish, and nearly featureless in reflected light, with an albedo of just 0.071, making it one of the least reflective bodies anywhere in the solar system. Those spectral properties resemble a class of asteroids called D-types, more commonly found much farther from the Sun, in the outer asteroid belt or beyond. If Phobos formed there and later fell into Mars's gravitational grip, that resemblance makes sense. The case against simple capture comes from orbital mechanics. Phobos and Deimos both circle Mars in orbits that are nearly circular and nearly aligned with the planet's equator  not the elongated, randomly tilted paths you would expect from a body randomly snagged out of a passing trajectory. Reproducing today's tidy orbits from a capture event requires invoking a long, complicated history of tidal reshaping that is difficult to confirm from the outside. Spectra point one way, dynamics point the other, and orbital observation alone has not been able to force a decision between them.

This is precisely the kind of impasse that laboratory chemistry can break and telescopes cannot. Scientists already have Martian material sitting in freezers and vaults on Earth, in the form of meteorites blasted off the planet's surface by ancient impacts and eventually landing here. Those meteorites carry a distinct isotopic signature  ratios of oxygen, chromium, and other elements that act almost like a fingerprint of Martian rock. If a Phobos sample turns out to share that fingerprint, it would be powerful evidence that the moon is built from Martian material, ejected into orbit by impact. If instead its isotopes resemble known carbonaceous or D-type asteroids, that points toward capture. Orbiting spectrometers can estimate surface mineralogy from a distance, but they cannot measure isotope ratios with the precision that mass spectrometers in terrestrial laboratories can achieve on an actual physical sample.

Understanding why Phobos matters at all requires sitting with just how strange a place it is. It measures roughly 27 by 22 by 18 kilometers along its three axes, an elongated, potato-shaped body rather than a sphere  it is too small for its own gravity to have pulled it into a round shape. One end is dominated by Stickney, an impact crater nearly 9 kilometers across, almost half the width of the moon itself, ringed by long parallel grooves whose origin is still debated: some researchers attribute them to boulders that rolled and bounced across the surface after the Stickney impact, others to stress fractures opened up by Mars's tidal pull. Surface gravity on Phobos measures about 0.0057 meters per second squared, something like 1,700 times weaker than what you feel standing on Earth. Escape velocity is only about 11 meters per second, or roughly 40 kilometers an hour  slower than a determined sprint. Phobos also orbits absurdly close to its planet, about 6,000 kilometers above the Martian surface, closer to its primary than any other known moon anywhere in the solar system, completing a full orbit in just 7 hours and 39 minutes. Mars itself takes about 24.6 hours to rotate once, so Phobos actually laps the planet faster than Mars spins beneath it  from the Martian surface, it rises in the west and sets in the east, crossing the sky twice in a single Martian day. That proximity comes at a cost: tidal forces are dragging Phobos inward by roughly two centimeters a year, and within thirty to fifty million years it will likely either break apart into a ring or crash into the planet it orbits. Deimos, smaller still and carrying only about a seventh of Phobos's mass, orbits much farther out, beyond the point where Mars's rotation and a moon's orbital period would match  so instead of spiraling inward, it is very slowly drifting away.

A body this small and this geologically inert has one enormous advantage as a scientific witness: it has essentially nothing to erase its own past. Mars has volcanoes, wind, and an atmosphere capable of resurfacing terrain over geological time. Phobos has none of that. Whatever material accumulated on and near its surface, whether captured asteroid or Martian ejecta, has likely sat there relatively undisturbed for a very long stretch of solar system history, altered mainly by the slow rain of cosmic rays and micrometeorite impacts rather than by any internal geological engine. A tiny, inactive moon can end up preserving evidence that a large, dynamic planet has long since erased from its own surface.

Assuming the October 2026 launch window holds, MMX will lift off from Tanegashima Space Center aboard an H3 rocket, beginning a cruise to Mars expected to take roughly a year, with arrival in the Martian system planned for 2027. Once there, the spacecraft settles into a quasi-satellite orbit around Phobos and spends about three years mapping and characterizing both moons using a suite of more than a dozen instruments  telescopic and multi-band imagers, a laser altimeter, an infrared spectrometer contributed by France, and the NASA-built gamma-ray and neutron spectrometer, MEGANE, which reads the elemental composition of Phobos's surface by detecting radiation naturally released when cosmic rays strike it. Only after this long reconnaissance phase does MMX attempt to touch the surface at all  landing on Phobos is difficult and unforgiving enough that mission planners want the terrain mapped down to individual boulders before committing.

Ahead of the mothercraft's own landing, a small rover named IDEFIX, built jointly by CNES and DLR, gets dropped first. IDEFIX weighs 25 kilograms, stands about 51 centimeters tall on four legs, and is released in free fall from an altitude of 40 to 100 meters above the surface. Because Phobos's gravity is so faint, that fall takes something like 60 to 80 seconds, and the rover touches down at under a meter per second  gently enough to survive without the elaborate landing gear a Mars rover would need. It then rights itself automatically, deploys its solar panels regardless of which way it landed, and spends roughly 100 days examining the ground directly with cameras, a radiometer, and a Raman spectrometer, which can identify minerals by how they scatter laser light. On a body where escape velocity is barely faster than a sprint, careless movement is a real hazard  a wheel spun too hard, or an ill-timed bounce, could in principle fling a rover clear off the surface. IDEFIX's low, careful mobility is designed entirely around that constraint, and its readings on how the regolith behaves mechanically will help mission controllers choose exactly where the exploration module itself should land.

Collecting the sample itself relies on two complementary systems. The first, called C-SMP, is a JAXA-built robotic arm tipped with a corer; after landing, it photographs the surrounding terrain, transmits that imagery to ground controllers on Earth, and then drills a few centimeters into a chosen patch of regolith, avoiding boulders and unusually hard ground. The second, P-SMP, was developed by Honeybee Robotics and works on an entirely different principle: it fires a burst of pressurized nitrogen gas at the surface to loft loose grains upward into a collector, sampling only the immediate top layer rather than drilling beneath it. P-SMP exists partly as insurance, in case Phobos's surface proves too hard for the corer, and partly because comparing its shallow samples with C-SMP's slightly deeper core lets scientists distinguish the most weathered surface dust from marginally fresher material just beneath it. Combined, the mission aims to bring home at least 10 grams of Phobos material  roughly two teaspoons.

Ten grams sounds small because it is small, but it is not a scientific afterthought. A returned sample is not limited by the mass, power, or fixed design of a single spacecraft instrument frozen years before launch. It can be divided among dozens of laboratories, archived for decades, and reexamined again and again with whatever the best available technique happens to be at the time  much the way lunar samples brought back by Apollo astronauts in the 1970s are still being analyzed today with instruments that did not exist when they were collected. Isotope-ratio mass spectrometry, electron microprobe analysis, and other laboratory techniques can resolve compositional differences down to the level of individual mineral grains, a resolution no orbiting spectrometer can match.

If the impact hypothesis turns out to be correct, MMX's sample return carries a pointed irony: a mission built to sample a moon, not the planet, may end up bringing home a piece of Mars itself  ancient crustal or mantle material, blown into orbit by a collision perhaps billions of years ago and never touched since. That would make MMX something genuinely distinct from NASA and ESA's separate effort to return samples cached on the Martian surface by the Perseverance rover; MMX is not collecting rock from the ground at Jezero Crater, but if the giant-impact hypothesis holds, it could still end up analyzing a fragment of Mars that has been orbiting overhead the entire time.

After sampling, MMX will lift off from Phobos, make a series of flybys past Deimos for additional observation, and depart the Mars system around 2030. The return module carries the samples on a roughly year-long cruise back to Earth, arriving around 2031, at which point a sample-return capsule separates and performs a high-speed direct atmospheric reentry, protected by a heat shield, before parachuting down for recovery  a technique with direct heritage from Japan's earlier Hayabusa asteroid sample-return missions.

The broader stakes reach past Phobos itself. Learning how to distinguish a captured body from an impact-formed one has implications for moons across the solar system, from Neptune's backward-orbiting Triton to the countless small satellites of the outer planets whose histories remain just as uncertain. And even a definitive answer for Phobos will come with a caveat. Roughly two centimeters of surface material have likely been churned, heated, and chemically altered by billions of years of micrometeorite bombardment and solar wind exposure  a process planetary scientists call space weathering. It is entirely possible that whatever isotopic signature Phobos carried at its birth has been partially scrambled by everything that has landed on it since, leaving the ten grams MMX brings home with a story that is real but incomplete: enough to narrow the debate between capture and impact, perhaps, without fully closing it.

YOU MAY ALSO LIKE

NASA's Nancy Grace Roman Space Telescope Is Chasing a Question Physics Can't Answer Yet

PLATO Mission: ESA's Search for How Many Earths Exist

Could Earth Life Survive on the Moon? Scientists Just Found a Clue