Why this research matters

Before we can build on the Moon, we need to practice on Earth — and that means finding rocks that genuinely resemble lunar soil. Only about 380 kg of real lunar material was brought back by the Apollo missions, and that is far too precious to crush inside a concrete mixer or feed into an oxygen-extraction reactor. Scientists therefore rely on lunar simulants: terrestrial materials that mimic the chemistry, mineralogy and grain-size of the real regolith closely enough to serve as stand-ins for engineering tests.

The problem is that most existing simulants are produced in the United States, supply is limited, and European alternatives remain scarce. This is where Mount Etna enters the picture. Europe's largest active volcano sits at a tectonic crossroads — the Ionian oceanic lithosphere subducting beneath the Calabrian Arc — and its complex magmatic plumbing has generated an unusually wide compositional range of volcanic products, from tholeiitic to alkali basalts. Some of those products turn out to be remarkably close to what Apollo astronauts scooped up on the lunar highlands.

The core question: Can we find, right on the slopes of Etna, a natural volcanic deposit whose chemistry, mineralogy and grain-size distribution are close enough to real lunar regolith to serve as a high-fidelity simulant — and can that same material actually be processed into construction bricks and breathable oxygen through In-Situ Resource Utilization (ISRU) techniques?

What we did

We collected samples from three geologically distinct sites on Etna, each representing a different volcanic environment that could mirror features we expect to find on the Moon:

2 600 m a.s.l.

Cisternazza Pit Crater

A summit pit crater mantled by fresh pyroclastic fall deposits from the paroxysmal eruptions of the last four decades — unweathered, glassy material ideal for simulating impact-generated lunar regolith.

CL-2 · CM-1 · CM-3
1 790 m a.s.l.

Monte Nunziata Lava Tube

Pyroclastic deposits behind the lining wall of an 1843-eruption lava tube — analogous to material that might accumulate inside lunar lava tubes, the prime candidates for future sheltered habitats.

MN-1
1 625 m a.s.l.

Tre Livelli Lava Tube System

Basaltic samples from different levels of a multi-tiered 1792–93 lava tube, capturing the internal variability of tube-hosted flows — relevant because the Moon hosts extensive lava tube networks.

TL-4 · TL-5

We then subjected these samples to a comprehensive analytical workflow inspired by — but not limited to — the NASA protocol for simulant characterisation. The pipeline combined bulk chemical screening (XRF + principal component analysis), quantitative mineralogy (Rietveld-refined XRD), cross-validation (ICP-OES), morphological analysis (optical microscopy, SEM-EDS, particle-size distribution), hyperspectral imaging in the VIS-NIR range, and two practical ISRU tests: alkali-activated material synthesis (i.e. "lunar concrete") and solid-gas carbothermal reduction for oxygen/water extraction.

Key findings

1. Chemistry: a near-twin of Apollo 14

Principal component analysis of major-element XRF data immediately flagged sample CL2 (Cisternazza pyroclastics) as the closest match to Apollo 14 highlands materials — specifically to sample 14259, a mature regolith collected 125 m west of the Apollo 14 Lunar Module on the Fra Mauro formation. The main compositional offset is higher alkali content (Na₂O ≈ 3.3 % vs 0.7 %; K₂O ≈ 1.8 % vs 0.5 %), which is expected: the Moon is depleted in volatile lithophile elements by a factor of 4–5 relative to Earth's primitive upper mantle.

Etna · CL2 ≈ Apollo 14 · 14259

OxideCL2 (Etna)14259 (Apollo 14)Match?
SiO₂47.5 %48.1 %✓ excellent
TiO₂1.9 %1.8 %✓ excellent
Al₂O₃16.4 %17.0 %✓ excellent
FeOtot11.7 %10.1 %✓ good
CaO10.8 %10.7 %✓ excellent
MgO5.5 %9.3 %△ lower
Na₂O3.3 %0.7 %△ higher (expected)
K₂O1.8 %0.5 %△ higher (expected)

2. Mineralogy: the right mineral cocktail, including 50 % glass

Rietveld-refined XRD showed CL2 carries a plagioclase-pyroxene-olivine assemblage with roughly 50 % amorphous (glass) phase — strikingly close to the 47.7 % glass measured in Apollo 14259. That glass fraction is crucial: on the Moon, impact melting and explosive volcanism produce abundant glassy phases in the regolith, and any serious simulant must reproduce this feature. Sample TL5, despite acceptable chemistry, was disqualified because it lacked glass entirely (>55 % feldspar instead).

Composite figure: optical and back-scattered electron microscopy of sample CL2, EDS spectra of its mineral and glass phases, and a MgO/Al2O3 versus CaO/Al2O3 diagram comparing CL2 to Apollo 14259 and to glasses in the lunar meteorite NWA 7948
Hand-sample and back-scattered SEM images of CL2 (a–f) show a mineral-and-glass texture that, at this scale, reads as lunar breccia. EDS spectra (g) resolve the amorphous glass phase from pyroxene, olivine, spinel and plagioclase. The MgO/Al₂O₃ vs CaO/Al₂O₃ diagram (h) places CL2's glass compositions inside the same field as glasses measured in Apollo 14259 and in the lunar meteorite NWA 7948. SEM-EDS · sample CL2

3. Spectral fingerprint: CL2 looks like lunar agglutinates

Hyperspectral analysis in the 0.4–2.5 µm range revealed that CL2's 1-µm absorption feature — diagnostic of Fe²⁺ in pyroxene and olivine — closely matches the spectral signature of the agglutinate-rich fraction of Apollo 14259. A PCA of spectral parameters (band centre, FWHM, asymmetry) confirmed this affinity. The slight ~30 nm offset in band minimum reflects the higher Ca content in Etna's augite compared with the low-Ca orthopyroxene dominant in the Apollo agglutinates.

What are agglutinates?
Agglutinates are clusters of mineral and rock fragments welded together by impact-generated glass. They can make up 60–70 % of a mature lunar soil and are a hallmark of prolonged space weathering. Their spectral resemblance to CL2 — a volcanic glass-rich pyroclastic — is expected because both consist of mineral fragments embedded in a glassy matrix.

4. Building with "Moon dust": up to 16.4 MPa compressive strength

Four alkali-activated formulations were prepared from CL2 powder and cured for 28 days at 25 °C and 65 % relative humidity. The best performer (sodium aluminate + sodium hydroxide, liquid/solid ratio 0.27) reached an average compressive strength of 16.40 ± 3.36 MPa. Adding 3 % urea slightly reduced peak strength (14.50 MPa) but dramatically improved batch-to-batch consistency (coefficient of variation dropping from 20.5 % to 6.9 %) — a critical advantage for automated robotic construction on the Moon.

On the Moon, where gravity is only 1/6 of Earth's, even a conservatively degraded material (accounting for vacuum and thermal cycling) would bear loads equivalent to a much stronger terrestrial concrete, making these values promising for habitats, landing pads and radiation shielding.

5. Oxygen and water extraction: comparable to certified simulants

CL2 was tested in a solid-gas carbothermal reduction plant at 1 100 °C for 8 hours using a H₂/CH₄ gas mixture. The CO extraction trend closely tracked those of certified highlands simulants NU-LHT-2M and LHS-1. CL2's higher ferrous-oxide content actually increased its reactivity with hydrogen, supporting faster reduction during hydrogen-only process steps — a practical advantage for ISRU oxygen production.

~50 % Amorphous glass in CL2, matching Apollo 14's 47.7 %
16.4 MPa Best compressive strength as alkali-activated material
r = 0.84 Pearson correlation of grain-size distribution with Apollo 14259
6 sites sampled across lava tubes and pit craters on Etna

Why Etna, and why now?

Mount Etna offers something no artificial blending plant can easily replicate: natural compositional variability. Its complex slab-edge geodynamics has produced both mare-like and highlands-like lithologies within a few kilometres of each other, and Sicilian law already classifies volcanic ash fallout as a reusable waste stream. That means tonnes of fresh simulant material can be sourced legally, sustainably and cheaply — right on the doorstep of European space agencies and research labs. As ISRU technologies mature and demand for large-volume testing grows, a readily accessible, high-fidelity natural simulant like CL2 could become an invaluable asset for the next phase of lunar exploration.


Full paper & media