May 8, 2026 · Tags: astronomy, JWST, exoplanets, geology, space
For the first time, we've measured the mineral composition of a planet's surface outside our solar system. Not its atmosphere. Not its size or density. The actual rocks.
The James Webb Space Telescope pointed its MIRI instrument at LHS 3844 b — a super-Earth 48.5 light-years away — and came back with a 5–12 micron thermal emission spectrum that reads like a geology lab report. The result, published in Nature Astronomy on May 4, 2026 by Sebastian Zieba, Laura Kreidberg, and colleagues: a dark, low-silica surface dominated by basalt or olivine-rich mantle rock, with no detectable atmosphere and no sign of volcanic outgassing.
"We see a dark, hot, barren rock, devoid of any atmosphere," said Kreidberg, Director at the Max Planck Institute for Astronomy.
It looks like a bigger, hotter Mercury.
The Planet #
LHS 3844 b is about 30% larger than Earth, discovered in 2018 by NASA's TESS mission. It orbits an M5 red dwarf star once every 11 hours — so close that it's tidally locked, with one hemisphere in permanent daylight and the other in eternal darkness. The dayside averages around 1,000 Kelvin (725°C / 1,340°F). The nightside is consistent with absolute zero.
That temperature contrast alone was the first clue. Spitzer Space Telescope data from 2019 showed the planet had the largest possible day-night temperature difference — no heat redistribution, no winds, no atmosphere thick enough to carry warmth around. A bare rock.
What Spitzer couldn't do was tell us what kind of rock. It had one broad wavelength band. JWST has a spectrograph.
How You Read Rocks 50 Light-Years Away #
The technique is called secondary eclipse spectroscopy. As the planet orbits behind its star, the combined light (star + planet) drops. That drop — 696 ± 18 ppm in this case — represents the planet's own thermal emission. By measuring that emission across wavelengths from 5 to 12 microns, you get a spectrum.
Different minerals emit infrared radiation differently. Silicate rocks have characteristic spectral features — the Christiansen feature, Restrahlen bands, Si-O stretching features — that shift depending on composition, crystal structure, and grain size. Granite looks different from basalt. Basalt looks different from olivine-rich mantle rock. Fine powder looks different from solid slab.
JWST's MIRI/LRS observed three eclipses, each spanning 2.58 hours, producing a 39-sigma detection of the planet's thermal emission. The team then compared this spectrum against two spectral libraries: a new, expanded library covering a wide range of rock types and textures (solid slabs, coarse crush, fine powder), and the RELAB database of measured mineral spectra from Earth, the Moon, and Mars.
What They Found #
Olivine clinopyroxenite — an ultramafic rock — fit the data best, at about 1-sigma. Basalt from the 1919 Kilauea lava flow fit at 1.5-sigma. Both are low-silica, magnesium-and-iron-rich rocks. Both point to a planet whose crust resembles Earth's oceanic basalt or, more likely, its mantle material.
Granite was ruled out at 8.9-sigma.
That's significant. Granite forms through plate tectonics and requires water. Its absence means LHS 3844 b almost certainly never had Earth-like geological recycling. As Zieba put it: "Since LHS 3844 b lacks such a silicate crust, one may conclude that Earth-like plate tectonics does not apply to this planet, or it is ineffective. This planet likely only contains little water."
Fresh powder surfaces were also ruled out — powders are too bright. But space weathering can darken them. Micrometeorite impacts and solar wind bombardment create nanophase iron particles and carbon deposits that darken regolith over time — the same process that makes the Moon's surface so dark. With just 0.5% nanophase iron mixed in, basalt powder fits at 1.9-sigma.
This leaves two scenarios:
- A young, geologically active surface — fresh basaltic rock with no time to weather, implying ongoing volcanism.
- An old, space-weathered surface — darkened regolith like the Moon or Mercury, implying prolonged inactivity.
The tiebreaker: volcanic outgassing. If the planet were actively volcanic, MIRI should have detected sulfur dioxide (SO₂) or carbon dioxide (CO₂). The data put tight limits on both — SO₂ below 10 microbars (3-sigma), CO₂ below 100 millibars (5-sigma). No volcanic gases detected.
The team favors the second scenario. LHS 3844 b is probably a dead, weathered rock. Mercury with better lighting.
Why This Matters #
This is the first time anyone has directly constrained the geological composition of a rocky exoplanet's surface. Not inferred from density. Not guessed from orbital parameters. Measured from emitted infrared light.
Previous exoplanet characterization focused almost entirely on atmospheres — transmission spectroscopy during transits, phase curves, emission photometry. That work has produced a growing catalog of atmospheric measurements, including the recent finding that TRAPPIST-1 b and c both lack thick atmospheres. But atmospheres are only half the story of a rocky world. The other half is what's underneath.
JWST has now opened a second door: exo-geology. A field that uses the same infrared spectral analysis planetary scientists have applied to the Moon, Mercury, Mars, and asteroids — but pointed at worlds we'll never visit.
The team has already secured additional JWST time. GO program 4008 (PI: Sebastian Zieba) obtained a phase curve of LHS 3844 b using NIRSpec G395H, covering 2.7–5.2 microns. This will constrain the surface composition at shorter wavelengths and, more importantly, measure thermal beaming effects — differences in how solid rock and powder emit radiation at different angles. The same technique used to study asteroid surfaces from spacecraft flybys, applied to an exoplanet.
A Cycle 4 proposal (GO 7953, PI: Kimberly Paragas) will observe eight more MIRI/LRS eclipses and one NIRSpec G395H eclipse, aiming to robustly detect surface spectral features at greater than 3-sigma — the first mineral-level identification on another world.
The Bigger Picture #
LHS 3844 b is not habitable. It's too hot, too airless, too close to its star. But the techniques used to study it are the same ones that will eventually characterize temperate rocky worlds.
M dwarf systems — the most common planetary hosts in the galaxy — are proving consistently hostile to atmospheres. TRAPPIST-1 b, TRAPPIST-1 c, LHS 3844 b, GJ 3473 b: JWST keeps finding bare rocks. The stellar wind and UV bombardment from M dwarfs, especially in their turbulent youth, appear to strip atmospheres efficiently. This has implications for the habitability of the billions of rocky planets orbiting these stars.
But even airless rocks tell stories. The absence of granite on LHS 3844 b tells us no plate tectonics. The absence of volcanic gases tells us no active volcanism. The presence of basalt or olivine tells us about mantle composition and the planet's formation history. A dead planet is still a geological object.
The emerging field of exo-geoscience — reviewed in a 2026 paper by a large collaboration — argues that connecting planetary interiors to detectable surface and atmospheric signatures is the next frontier. What JWST has done for LHS 3844 b is proof of concept. Different rock types have distinguishable spectra. Space weathering has measurable effects. Surface texture (slab vs. powder) leaves detectable imprints.
Future telescopes will extend this to cooler, more interesting worlds. NASA's Habitable Worlds Observatory, designed for direct imaging of Earth-like planets around Sun-like stars, will eventually provide the contrast and spectral resolution to read the surfaces of planets in habitable zones. When that happens, the baseline established by JWST on hot, dead rocks like LHS 3844 b will be the calibration set.
For now, 50 light-years away, a dark basaltic world rotates in lockstep around its dim red star. No atmosphere. No water. No plate tectonics. A larger Mercury, baking at 725°C on its permanent dayside, frozen on its nightside.
And we know what its rocks are made of. That's new.
Sources: Zieba, S., Kreidberg, L., et al. "The dark and featureless surface of rocky exoplanet LHS 3844 b from JWST mid-infrared spectroscopy." Nature Astronomy (May 4, 2026). Kreidberg, L. et al. "Absence of a thick atmosphere on the terrestrial exoplanet LHS 3844b." Nature 573, 87–90 (2019). Center for Astrophysics | Harvard & Smithsonian press release (May 4, 2026). Max Planck Institute for Astronomy press release (March 2026). NASA/JPL.