Subsystem 10 of 12 · Construction
Lunar Habitat ACT, the Autonomous Construction Testbed·Construction·Concept definition, revision A·August 2026·See it in the tour

Building the Moon out of the Moon

The reason the outpost exists. A laser fuses the ground into something that behaves like ceramic, and every structure built makes the next delivery cheaper.

11 m²of slab per day
$57.2 MNASA award to ICON
0 kgbinder shipped
Contents
  1. Why the pad comes first
  2. Laser vitrification
  3. What gets built, and in what order
  4. Rate and energy
  5. Material properties, honestly
  6. What is not yet known
  7. Increment plan
  8. Sources and further reading
Section 01

Why the pad comes first

A lander touching down on unprepared regolith is the single most destructive event that can happen to a lunar base. The exhaust plume excavates the surface and throws material outward at speeds the plume models put above two kilometres per second for the finest grains, and with no atmosphere to slow it, that material travels ballistically until it hits something.

Apollo 12 landed 155 metres from the Surveyor 3 probe and sandblasted it badly enough to pit the camera housing and strip paint; the pits on the parts the crew brought home bound the impact speed between about 300 metres a second and two kilometres a second. That was a small lander. A cargo vehicle delivering the next habitat module is very much larger.

So the sequence is forced: survey, grade, sinter a pad, raise a berm, and only then land anything else nearby. Everything in the construction programme follows from that ordering.

155 mApollo 12 to Surveyor 3
2 km/sejecta velocity
0atmosphere to slow it
65 mpad to habitat separation here
Section 02

Laser vitrification

Lunar regolith is largely silicates and metal oxides. Heat it past roughly 1,200 °C and it does not burn, it melts, and on cooling it forms a glass ceramic with useful compressive strength. In laser tests on a south pole simulant, sintering begins near 1,180 °C and a full melt needs about 1,360 °C, and melting proved about ten times more energy efficient per unit of consolidated material than sintering, which is why the head here melts rather than sinters. No cement, no water, no binder, nothing shipped from Earth.

This is the process NASA's MMPACT programme (Moon to Mars Planetary Autonomous Construction Technology) and ICON are developing under a 57.2 million dollar award, marketed as Laser Vitreous Multi material Transformation. It is the most credible surface construction technique currently funded.

Grade and compact

A blade flattens the site and a roller closes the voids. Sintering an uncompacted surface leaves a crust over cavities, which cracks the first time a lander puts weight on it. This step is skipped by nobody twice.

Melt with a directed beam

A high power laser on the gantry raises the top 30 to 50 mm past 1,200 °C. The head tracks back and forth laying one course at a time.

Cool by radiation alone

In vacuum there is no convection, so the melt sheds heat only by radiating and by conducting into the soil beneath. Cooling is slow and even, which is exactly what gives the slab its strength. Rushing it introduces thermal stress and the slab fails on the first landing.

Inspect

An ultrasonic head checks the course for voids before another goes on top. Cheap now, impossible later.

Index and repeat

The gantry steps one course forward. Eleven square metres of finished slab a day.

Section 03

What gets built, and in what order

StructureIncrementArea or massRateWhy it is at this point in the order
Graded pad bedsite prep1380 m²8 daysNothing can land safely until this exists
Sintered apron22 m diameter1380 m²35 daysThe slab itself
Module skirtscompacted regolith to a fifth of hull height1not yet budgetednot yet budgetedAnchors each hull and protects its foot once the modules are down; the only regolith against the vessels in the built increment
Blast berm29 m outer21,400 pavers36 daysCatches what the apron does not
Module shielding2 m over five vessels, planned21,850 t310 daysThe single largest earthmoving job on the site, not yet started in the built increment
Graded roadsfour routes32,100 m²190 daysCuts dust generation and rover wear
Equipment padsplant and array3260 m²24 daysStops the plant settling
Printed shell trialtest article418 m²60 daysThe step that decides whether volume is launched or made

Module shielding will dominate everything once it starts. Eighteen hundred tonnes of regolith moved by a machine handling six tonnes a day is most of a year of continuous operation, and it is why the excavator is the busiest thing on the site and why autonomy matters so much. The site is shown before that job: the vessels stand on their skirts and the cover is still to come.

Section 04

Rate and energy

Construction output per day at full duty Sintered slab 11 per day m², 9 kW while the beam is on Pavers cast 40 per day units, 7 kW Bulk regolith placed 6 per day tonnes, 4 kW Road surface 8.5 per day m², thinner course than the pad Excavated and screened 14 per day kg/hr to the ISRU plant, shared machine Total construction draw is about 9 kW average, peaking near 16 kW when the gantry and excavator run together.
These rates assume the machines run through the lunar night on fission power. Restricting construction to daylight would roughly halve the annual output and push the shielding job past two years, which is the practical argument for the reactor.
Section 05

Material properties, honestly

Sintered regolith is not concrete and should not be described as though it were. It is a glass ceramic: strong in compression, weak in tension, and brittle.

PropertySintered regolithPortland concreteImplication
Compressive strength20 to 45 MPa20 to 40 MPaComparable. Good for pads and footings.
Tensile strength2 to 4 MPa2 to 5 MPaBoth are weak. Neither spans without reinforcement.
Density2.4 to 2.8 g/cm³2.4 g/cm³Similar
Thermal shock resistancePoorModerateThe reason cooling cannot be rushed
Reinforcement availableISRU iron alloy (in situ resource utilisation)SteelThe alloy from the plant is the rebar
The limit worth being clear about

A pressurised shell needs tensile capacity, and sintered regolith does not have it. The increment four trial is a shell with ISRU iron reinforcement, and whether that works is genuinely unknown. Everything up to increment three, which is pads, berms, shielding and roads, is compression only and well inside what the material can do.

Section 06

What is not yet known

Nothing in this document has been done on the Moon. The MMPACT and ICON work is funded development on Earth; the November 2022 award pays for the construction system, and no slab has yet been laid on the surface. The 11 square metres a day is an allocation for a gantry that does not exist, and the strength table comes from simulant tests, not from lunar regolith, with recent cryogenic cycling tests showing sintered simulant losing strength over repeated cold soaks. The excavator rate is the largest assumption in the build order: six tonnes a day is an allocation, while NASA's ISRU Pilot Excavator (IPEx) demonstration target is ten tonnes over eleven days, under a tonne a day, for a machine of the class the site shows, so the 310 day shielding job assumes either a larger machine or several of them, and neither has flown. The 65 metre pad distance rests on the sintered apron and the berm catching the low angle fan; the only measured ejecta case is Apollo 12 at 155 metres, and the landers this base expects are far larger than the lunar module. The compacted skirts at the foot of each hull are not yet budgeted for mass or placement time. A pressurised shell needs tensile capacity and sintered regolith does not have it; whether a shell reinforced with ISRU iron works is genuinely unknown, and that is the increment four trial. Everything before it is compression only and inside what the material can do.

Section 07

Increment plan

Increment 1, survey, grade and pad

The site is surveyed on composition and slope, graded flat, and the apron laid. Nothing else lands until it exists. Once the modules are down, a compacted regolith skirt is built against the foot of each hull to about a fifth of its height.

Increment 2, berms and shielding

The blast berm around the pad, then two metres of regolith over every pressure vessel. This is the longest single job in the programme, and the base is shown before it: each hull stands clear on its skirt.

Increment 3, roads and equipment pads

The four routes, and footings for the plant and the array. Dust generation across the site falls sharply once these are down.

Increment 4, printed pressure shell

A reinforced test article, pressurised and instrumented but not crewed. If it holds, habitat volume stops being launched and starts being made.

Sources

Sources and further reading

Every figure in this document traces to one of these. Agency documents and peer reviewed work first, reporting only where it is the primary record.

ICON, ICON to develop lunar surface construction system with 57.2 million dollar NASA award, November 2022 iconbuild.comThe Project Olympus award this design assumes
NASA, NASA enables construction technology for Moon and Mars exploration, the MMPACT project nasa.govProgramme context and process maturity
NASA, Analysis of Surveyor 3 material and photographs returned by Apollo 12, NASA SP 284, 1972 https://ntrs.nasa.gov/citations/19720019081The sandblasting record that justifies building the pad first
Immer et al., Apollo 12 lunar module exhaust plume impingement on Lunar Surveyor III, Icarus 211, 2011 https://www.sciencedirect.com/science/article/abs/pii/S001910351000432XSurveyor 3 pitted 155 m from the landing by particles at 70 m/s or faster
NASA, The behaviour of high velocity dust generated by lunar landers, AAS guidance and control conference, 2020 https://ntrs.nasa.gov/citations/20205003594Plume ejecta above 2 km/s for the finest fraction; the Surveyor pitting bounds of 300 to 2,000 m/s
Laser melting versus laser sintering: large area heat processing of lunar south pole simulant, 2025 https://www.sciencedirect.com/science/article/pii/S295043172500036XSintering from about 1,180 C, full melt above 1,360 C, melting about ten times more energy efficient per unit consolidated
Synthetic space bricks from lunar and Martian regolith via sintering, Advances in Space Research, 2024 https://www.sciencedirect.com/science/article/abs/pii/S0273117724006227Compressive strength up to 45 MPa in furnace sintered simulant
A structural assessment of unrefined sintered lunar regolith simulant, Acta Astronautica, 2018 https://www.sciencedirect.com/science/article/abs/pii/S0094576517305404Compressive strength far above tensile; tensile stress governs the safety of anything built from it
Mechanical deterioration of sintered lunar regolith simulants under extreme cryogenic thermal cycling, Science China Technological Sciences, 2025 link.springer.comStrength loss over repeated cold soaks; the thermal shock line in the property table
NASA, ISRU Pilot Excavator technology readiness level 5 design overview, AIAA ASCEND, 2024 arc.aiaa.orgThe 10 tonne, 11 day excavation target behind the excavator caveat