ISRU: In Situ Resource Utilization, making use of local materials
Bootstrapping: landing a cargo consisting of machines that use local materials to build more machines, habitat, produce oxygen, etc.
Building it on the Moon
The first thing that must be done before the construction of a lunar industrial facility is determining the best place to locate it. Polar locations offer craters containing ice and prolonged sunshine Preferably, the location for the first industrial settlement will have plenty of flat ground with very few boulders. A mare/highlands "coast" might be ideal. Ilmenite, basaltic mare regolith, KREEP, pyroclastic glass deposits, highland regolith and lava tubes should be reasonably nearby. Polar ices should also be accessible. A location in Mare Frigoris might be best.. According to Peter Kokh:
”In the north, Mare Frigoris offers coastal areas less than half the distance from the north pole than southern coastal regions (Mare Australe, Mare Nubium, or Mare Humorum) are from the south pole. Further, Mare Frigoris is in the "Imbrium Fringe" area that Lunar Prospector has shown to be thorium-enriched.
Previously, this author had been partial to a settlement in Mare Crisium... But Mare Frigoris comes in a close second in this regard. Here, just north of the crater Plato and the Alpine Valley (providing access to Mare Imbrium and the whole "chain" of nearside maria) might be an especially propitious place to set up an initial settlement.”
Highland regolith is richer in calcium and aluminum and mare regolith is richer in iron, titanium and magnesium. Locating at the edge of the mare will allow access to both kinds of regolith while polar locations are all in highland areas. The downside of a mare location is that darkness will last for two weeks at a time and power storage systems will be needed to maintain life support and protect equipment from the cold. The first payload to be landed would be a LUNOX production plant as discussed above to load reusable landers.
2. Site Preparation
Once a site is selected it needs to be prepared. The site must be leveled and small craters must be filled in. Boulders must be dynamited and the rocks pushed aside. Markers must be placed to indicate the locations of solar panel farms, landing pads, roads, walkways, a warehouse, a pad for production machinery and inflatable habitat modules. Robotic bulldozers and graders will be called for. There must also be a solar panel farm and wiring systems to recharge the batteries in the bulldozers and graders. These machines might be powered by tethers or microwave beams from the solar panel farm. Receiving antennas on the machines will just be low mass wire meshes with some Zener diodes and this will not burden the 'dozers and graders. This would free the machines from the burden of heavy battery or fuel cell packs and the need to shut down and recharge for several hours at a time. Robots that can drill holes in boulders and place explosives will also be needed. Logically, all the robots will retreat to extreme range when boulders are blasted.
Clearly, the first payloads to the Moon, after the LUNOX production plant, must be solar panels and associated hardware along with several bulldozers and graders. Robots to deploy the solar panels and wiring systems will also be needed. To protect the machines during nightspan it might be desirable to have infrared lamps to keep parked machines warm and power storage systems to energize the lamps. Batteries, flywheels or fuel cells come to mind.
Fuel cell systems will require insulated tanks to store liquid hydrogen and liquid oxygen, plumbing systems, water electrolysis systems, and refrigeration devices to liquefy hydrogen and oxygen. That sounds like a complicated mess when compared to batteries or flywheels; however, there is an important advantage to the use of fuel cells for nightspan power storage. Fuel cell systems can augment rocket fuel storage facilities with their cryogenic reactant storage tanks and liquification systems.
The LUNOX plant will have its own small robot regolith loaders. More mining capacity may be desired. It should be possible to equip bulldozers with mining shovels so that these can do two jobs instead of just one. Attempting to define the components of a lunar “industrial seed,” we can imagine the first payloads to the Moon will include but not be limited to:
- LUNOX production plant
- general purpose teleoperated robots
- solar panels, supports, motors, reflectors, wiring, switches, invertors, etc.
- power storage systems, probably fuel cell systems complete with insulated tanks for cryogens, piping, pumps, valves, electrolysis and refrigeration systems that can double as a rocket propellant depot
- microwave transmitters and/or tethers
- bulldozers and graders that can also excavate
- IR lamp systems
- oxygen generators (e.g. vapor pyrolysis or magma electrolysis)
- pumps etc. for “gassing up” lander rockets
3. Early Development
Once the site is leveled out and large rocks removed, it will be developed. Landing rockets will cause dust to spray all over and dust could damage machinery especially if it gets into bearings. Dust sprays could disaffect solar panels also. Several landing pads will be made. Wheeled robots with microwave generators could sinter or melt the basaltic ground to a depth of several inches (5 to 10 cm) at least. Bulldozers could berm up regolith around the pads. With three pads one rocket could be lifting off while another lands and a third one is waiting for service. Landers or "Moon Shuttles" might have wheels on their landing legs so they can be towed off the pad. The landing pads should be fairly big. A diameter of one hundred meters will allow a large margin of safety if a rocket is a bit off course. The pads would be located about a kilometer away from the habitat so that the chance of a Moon Shuttle rocket going off course and crashing into the habitat and killing everyone is very low. Roads from the landing/launch pads to the habitat and work area will be paved with microwaves.
Lunar workers, their machines and robots will need a nice hard floor made by microwave roasting of the basalt as is done for the landing pads and roads to mount production machines on. Plain old ground is no good. Spacesuited human workers and robots would kick up dust and some machines like power forging hammers would pound or vibrate into the dusty surface. The floor might be thicker than the pads and roads. A large foil or aluminized Mylar parasol to shield machines and workers from the hot sun could be erected and teleoperated robots could work the production machines. Now and then humans will have to go outside in turtleback spacesuits to do some work. There will be microwaved walkways from the habitat modules to the production machine area. There will also be a warehouse consisting of a microwaved pad with a parasol to store cargo containers as they arrive by Moon Shuttle.
In addition to solar panel farms there must be power storage for nightspan not only to keep machines warm with IR lamps but to power lights, radios, computers and mechanical life support systems in habitat modules. A small nuclear generator would help. The microwaved basalt pads will serve as "thermal wadis" and cool slowly after sunset. That will be easier on the machines. Sudden thermal shock can crack metals. Even in polar locations there will be periods of darkness but these will last only a couple of days while in lower latitudes where the mare are darkness will prevail for two weeks out of every month. In the distant future there could be a solar power satellite at EML1 and a circumlunar power grid with solar panel farms around the Moon to supply full power at all times.
Secondary payloads to the Moon will include but not be limited to:
- more solar panels, wiring systems, power storage, possibly a small nuclear generator
- at least two rovers, preferably more, in case one breaks down with microwave generators to make pads and roads
- inflatable habitat with mechanical life support systems and some tanks of oxygen to inflate the habitat
- parasols with support poles to protect workers and equipment from solar heat and to prevent radiation of heat from equipment by night
- running lights, flood lamps and radio antennas.
- supplies of dehydrated and freeze dried foods, drinking water and medicines, clothing, bedding, towels and wash cloths, light weight furniture, recreational supplies (dart board, chess set, playing cards, board games, dice etc.), toiletries and sundry items (toilet paper, toothpaste, brushes, razors, blades, soap, lotion, shaving cream, etc deodorant, cologne, and perfume will have to wait until lots of air cleaning vegetation is cultivated within habitat  )
At least this much should be in place before human crews move in and start working. The bulldozers and graders with shovel attachments must cover the inflatable habitat with six meters of regolith for radiation, thermal and micrometeoroid protection. Protection from galactic cosmic rays in free space would require about 11 tons of regolith per square meter of hull to reduce radiation dosage to 20 mSv/yr and 6.6 mGy/yr. On the lunar surface, the Moon blocks out half of this so 9 tons per square meter would be needed . Since bulk regolith is about 1.5 times as dense as water 6 meters of regolith would suffice  Without shielding humans could make only brief sorties on the Moon. It is foreseeable that robots might experience glitches that halt the project and humans become necessary to get things going again. Space workers could land and stay inside their spacecraft for a few days until they get the machines back up and running.
The bulldozers and their shovels are just the beginning. Massive amounts of regolith must be moved to support a serious Moon mining operation with the goal of building mass drivers, solar power satellites and other constructions in space. A slusher system seems best . This consists of a bucket attached to some steel cables. A winch pulls the bucket through the dust and it picks up a load. The load is lifted and dumped into a truck or an ore car on rails. A second set of cables wrapped around some pylons with pulleys at the edge of the excavation is pulled on by the motorized winch and the empty bucket is dragged out and readied to scoop up another load of regolith. This will be more efficient than making excavators scoop up a load, carry the load and their own weight to the refinery, dump the load and drive back to the hole, and repeat the process. The slusher can work continuously. At first trucks will haul lunar regolith to the refinery on microwaved basalt roads perhaps. Later on, a railway system will be constructed. Cars riding on steel rails will endure much less rolling friction and that will save energy. They won't kick up dust either. When the slusher has dug up a pit and can dig no more it can be relocated. Rail systems can be extended to the new dig site. Slushers will be set up in different locations to dig mare regolith and highland regolith. Mare regolith is richer in iron, magnesium and titanium and can be melted down and cast as is. Highland regolith is richer in calcium and aluminum. These “soils” would be processed differently depending on what substances are desired. The Moon Shuttles will land a few more payloads at this time:
- slusher systems consisting of cables, buckets, motorized winches, pylons
- hauling trucks
Fig. 3 Side view of slusher mining system. NASA
Fig. 4 Open pit mine with slusher system. NASA
used with permission from Space Studies Institute
Fig. 5 Regolith composition
5. Materials Production
There are many proposals for the extraction of materials from lunar regolith. The regolith is rich with oxygen, silicon, iron, calcium, aluminum, magnesium, titanium and has significant traces of manganese, chromium, sodium, potassium and phosphorus. Even without complex electrochemical systems for extracting these metals there are resources of great value. Mare regolith is basaltic. It can be dug up, melted in a solar or electrical furnace, and crude castings can be made in molds dug into the ground and finer castings can be made in iron molds. It can also be sintered instead of cast. Sintering means that the material is compacted into molds and heated only enough for the edges of its particles to fuse together. This can make worthy items like bricks, blocks, tiles, slabs and rods without as much energy as full melting and casting requires. Basalt can also be melted and drawn through platinum-rhodium bushings to make fibers. A wide variety of things, perhaps even habitat modules and solar power satellite frames, could be made from basalt and glass. Metals will still be needed for vehicles, digging machines, railways, spacecraft, electrical systems, electric motors, etc.
Iron molds sound like heavy cargoes to import to the Moon. Perhaps they could be made on the Moon in large numbers. Magma electrolysis yields ferrosilicon and silicate ceramic as well as oxygen . Ceramic blocks could be cast in molds dug in the ground. It might be possible to perform serial magma electrolysis in which case iron could be derived separately from silicon. This iron could be powdered and fed into 3D printers that use electron beams or lasers to fuse metal layer by layer to make all sorts of shapes. If serial magma electrolysis is not possible there is another resource of great value on hand--meteoric iron-nickel fines that are present in regolith all over the Moon at concentrations of a few tenths of a percent by mass. These could be harvested by rovers that have low intensity magnetic separators. This could be the first metal produced on the Moon. The particles are fused with silicates and can be purified by running them through centrifugal grinders to shatter the brittle silicates followed by another magnetic separation. In 1981, Dr. William Agosto projected that this system could produce 552 tons of a 99% pure iron/nickel feedstock annually . After sieving and sizing the powder could be placed in 3D printers to make iron molds of various sizes and shapes for casting and sintering basalt. A third way to obtain iron involves the roasting of regolith at 1200 C. in the vacuum to drive off FeO, condensing the iron oxide and reducing it with hot hydrogen or using electrolysis to free up the iron and obtain oxygen .
Additive manufacturing,”3D printing,” with metals is commonplace today. Stainless steel, low alloy steel, maraging steel, cobalt and nickel alloys are all used presently . Direct metal laser sintering can produce solid parts without using a binder and it can make parts with complex geometries that CNC milling cannot . Planetary Resources and 3D Systems actually printed up a model spacecraft with powdered meteoric iron-nickel material . Certainly, lunar meteoric iron-nickel particles can also be used for additive manufacturing with electron beams or lasers.
Basalt could be a very important base material. It is harder than steel and abrasion resistant. It is strong in compression but not so strong in tension and it is rather brittle. Uses for basalt include [from reference 19]:
machine base supports (lathes, milling machines), furnace lining for resources extraction operations, large tool beds, crusher jaws, pipes and conduits, conveyor material (pneumatic, hydraulic, sliding), linings for ball, tube or pug mills, flue ducts, ventilators, cyclers, drains, mixers, tanks, electrolyzers, and mineral dressing equipment, tiles and bricks, sidings, expendable ablative hull material (possibly composited with spun basalt), track rails, "railroad" ties, pylons, heavy duty containers for "agricultural" use, radar dish or mirror frames, thermal rods or heat pipes housings, supports and backing for solar collectors
nozzles, tubing, wire-drawing dies, ball bearings, wheels, low torque fasteners, studs, furniture and utensils, low load axles, scientific equipment, frames and yokes, light tools, light duty containers and flasks for laboratory use, pump housings, filters/partial plugs
Spun basalt (fibers)
cloth and bedding, resilient shock absorbing pads, acoustic insulation, thermal insulation, insulator for prevention of cold welding of metals, filler in sintered "soil" cement, fine springs, packing material, strainers or filters for industrial or agricultural use, electrical insulation, ropes for cables (with coatings)
Fig. 6) Basalt fibers can be made the same way glass fibers are made.
Meteoric iron/nickel fines can be used for more than making molds for casting or sintering basalt. They contain 5 to 10% nickel, 0.2% cobalt and traces of germanium, gallium and platinum group metals (PGMs). Iron, nickel and cobalt can be separated by treating the fines with carbon monoxide gas. High temperature vaporization, ionization and electrostatic separation might also be applied. Nickel and PGMs have catalytic properties. Nickel can make steel harder and stronger without making it more brittle. Cobalt can by used for high speed drill bits and cutting tools. It can also stain glass a deep blue. Germanium and gallium can be used in electronics and photocells.
There are also traces of solar wind implanted volatiles (SWIVs) in regolith. Significant quantities of hydrogen, helium, nitrogen, water, carbon monoxide, carbon dioxide and methane can be obtained by heating regolith up to about 700 C. . At higher temperatures sulfur, potassium and sodium will also be liberated . Teleoperated machines that plow through the relatively smooth mare with bucket wheel loaders could roast out these elements in an onboard furnace and store the substances in tanks . The machines would only return to base when their tanks were full and heavy stationary refrigeration equipment could separate the gases and liquids. Hydrogen could be combined with oxygen to make water for life support systems and gardens. Nitrogen would be very important for fertilizer. Carbon could be used to make steel and add CO2 to atmospheres that support plant life.
Steel seems to be an unlikely material on the Moon where only small amounts of carbon exist. In reality, a tiny amount of carbon makes a large quanitity of steel. Mild steel is 0.05% to 0.35% carbon. Alloyed with some nickel very high quality steels can be made. There will be no roaring coke filled blast furnaces or basic oxygen furnaces sending out showers of sparks on the Moon. Steel could be produced by the ancient crucible steel process. Iron powders, rods or plates would be packed with carbon powder and brought up to red heat in a furnace made of basalt or a ceramic made on the Moon such as the spinel rich ceramic produced by magma electrolysis for about a week. The carbon will dissolve into the iron and form steel. The steel and carbon could be magnetically separated and the steel could be homogenized by melting to disperse the carbon evenly throughout the metal. During this melting the steel could be mixed with calcium aluminate flux to remove impurities. The CaAl2O4 flux could be produced by roasting highland anorthite at 2000 C .
Highland regolith contains less iron and magnesium than mare regolith but it is richer in calcium and aluminum. It can make a ceramic that does not melt until 1500 C. unlike basalt that melts at about 1250 C. Roasting highland regolith at up to 2000 C. can drive off silicon dioxide and enrich the calcium and aluminum oxide components to make hydraulic cement. If anorthite is extracted by electrostatic separation and roasted at 2000 C. and hotter calcium aluminate can be obtained . A furnace that can handle temperatures this high might consist of a slip cast aluminum oxide or silicon carbide crucible. Concentrated solar energy or an electron beam would heat the material in the crucible. Only part of the charge would be heated while the rest of the material, in this case anorthite, serves as insulation. Robots will dig out roasted material that has released SiO2 and formed calcium aluminate (CaAl2O4). Electrochemical processing of CaAl2O4 can yield aluminum and calcium metals . Calcium is an excellent electrical conductor. It might be possible to condense the SiO2 released and use it for glass making. If the furnace crucible can have active cooling achieved by drilling passages in its walls through which an inert gas is pumped that dumps its heat into the cold of outer space from shielded radiators it might be possible to construct more effective high temperature furnaces on the Moon.
We can see that additional payloads to the Moon should include but not be limited to:
- solar or electrical furnaces for melting and pouring basalt
- small digging tool attachments for making crude sand molds in the ground
- some iron starter molds for basalt
- platinum-rhodium bushings and whatnot for basalt fiber drawing
- heaters to sinter basalt packed into iron molds, packing tools for robots
- heaters, perhaps induction heaters, to melt steel
- magma electrolysis cells
- metal powdering equipment (centrifugal electric arc perhaps)
- rovers with low intensity magnetic separators for havesting meteoric iron fines
- centrifugal grinders
- 3D printers that can make heavy iron molds
Fig. 7) Powdering metals for rocket fuel and 3D printing will require special equipment.
- carbon monoxide processing equipment
- rovers for harvesting solar wind implanted volatiles
- cryonic refrigeration/distillation equipment
- electrostatic separation devices
- furnaces for roasting anorthite at 2000 C. +
- electrochemical equipment for aluminum and calcium extraction
- lithium fluoride and calcium fluoride electrolyte
- inconsumable cermet electrodes
With this equipment it should be possible to produce iron molds for basalt and furnaces with basalt, anorthite or spinel rich ceramic linings for steel. Nickel, cobalt, small amounts of germanium, gallium and PGMs, hydrogen, helium, nitrogen, carbon, sulfur, iron, aluminum and calcium also become available. Two more metals can also be had --magnesium and titanium. Ferrosilicon from magma electrolysis can be mixed with magnesium oxide obtained by roasting mare regolith at 1500 C.+. The mixture can be heated to 1200 C. under vacuum conditions and magnesium metal will boil out and can be condensed. Titanium can be obtained by mining in ilmenite rich mare regolith. The ilmenite can be concentrated with electrostatic separators . A fluidized bed can be made of welded steel plates and pipes and possibly some basalt parts in which the ilmenite is treated with hydrogen gas at 1100 C. Water and fused particles of titanium dioxide and iron will form. The water will be electrolyzed to recover hydrogen and gain oxygen . The TiO2 and iron particles must be separated possibly by treatment with CO gas to form iron carbonyls . The titanium dioxide could then be electrolyzed in FFC cells with inconsumable electrodes to obtain sponge titanium metal. This would be melted in a high temperature furnace at over 1800 C. or powdered for 3D printing.
Besides fluidized beds for ilmenite processing, metal plates both flat and curved and metal pipes will be needed to build pressurized cabins for ground vehicles, railroad cars and spacecraft, habitat modules, rocket propellant tanks, liquid and gas storage tanks, and numerous other products.
This would call for payloads of:
- rolling mills for making metal plates
- centrifugal casting machine to make basalt pipes
- extruder to make metal pipes
- accessory devices for making magnesium and titanium extraction devices
- FFC cells with calcium chloride electrolyte
However, rolling mills and extruders can be very massive pieces of equipment. A rolling mill could amass 40,000 kilograms or more and an extruder up to 10,000 kilograms or more, although there are smaller lighter versions even bench top scale rolling mills and extruders. A Falcon Heavy rocket could put 54,400 kilograms in LEO, so the problem is not getting it up there. If 1000 metric tons of cargo was lifted to the Moon a 40 metric ton rolling mill would only be about four percent of the total. Several pieces of heavy equipment besides a rolling mill like extruders, a large engine lathe and forging presses could be sent to the Moon and there would still be plenty of space left over for other things if 1000 tons is the mark. The question is, might it not be better to send the machinery for making heavy equipment on the Moon? Then make several copies of the heavy equipment on the Moon? How would this be done? What are these machines for?
6. Metal Massive, Unitary, Simple Things
It is not always necessary to melt and cast metals. That requires lots of energy, time and labor by men and machines. Cold metals can be shaped by rolling, extrusion and spinning. This work hardens the metal. When this is not wanted, metals can be hot worked. It takes energy to heat the metal but not as much as melting does and less horsepower is needed to roll or extrude the softer hot metal.
Foils: food wrapping, parasols, solar shields, reflectors to increase solar panel output
Sheets: buckets, bins, tool holders, shelves, drawers, computer and electronics casings, tableware
Flat Plates: slusher buckets or scrapers, excavator buckets, bulldozer and road scraper blades, ground vehicle parts, spacecraft frames, metal floors, fluidized beds (with some tubes and other parts), appliance parts and casings, even pots and pans by stamping small circular thin metal plates.
Curved Plates and Spun Domes: rocket propellant tanks, fuel cell reactant tanks, water tanks, oxygen (and other gases) tanks, pressurized ground and space vehicle cabins, parabolic solar trough reflectors, spun domes for radio and solar concentrator dishes.
Rails, Bars, Beams: ground vehicle and earth moving equipment frames as well as other parts, building support structures, railroad tracks.
Rods: axels, "tent" or canopy poles, radio antennas, rebar, "telephone" poles for power lines and phone lines.
Wires: power lines, phone lines, electrical wiring, motor coils, steel cables for earth moving equipment.
It can be plainly seen that the simple objects made by rolling, extrusion, spinning and drawing have many uses. The heavy equipment needed for this can mass produce these items from lunar metals. Aluminum, pure iron, meteoric iron-nickel and steel will be most useful. Magnesium is soft but not very ductile unless it's hot. Titanium is hard to cold work but can be hot worked. Many titanium parts could be made from powder by electron beam fusing or sintering, a kind of 3D printing, outside in the vacuum.
7. Casting on the Moon
There can be no doubt that materials production devices from meteoric iron harvesting rovers to solar and electrical furnaces and other things including solar panels for power must be sent to the Moon ready to work and supply materials or nothing can happen at all. Some production devices like 3D printers must be sent up ready to work also.
Working with metals will require furnaces for melting and heat treating and heavy machines like rolling mills to work the metals into useful things. Meteoric iron containing 5% to 10% nickel and pure iron from other sources could be converted to steel by roasting it with carbon. This is the ancient crucible steel process. The Moon doesn't have a lot of carbon but a tiny amount of carbon can make a large quantity of steel. Casting that steel into large rollers a meter or more (40 inches) in diameter and two meters (80 inches) long that are later ground and polished with CNC (computerized numerical control) machines to within two ten thousandths of an inch presents a problem. Fairly pure silica or olivine sand will be needed for expendable molds along with clay and water to bind the mold. While sand might come from regolith, clay is non-existent on the Moon and water is precious. In the vacuum the water will sublime and the wet mold will dry up. Unless this job is done inside a pressurized structure so that the water can be recovered from the air within that water will be lost.
What about using binders other than clay? Sand molds can be made with resin but the Moon lacks the light elements needed to make resin so it would be imported and recycled. The resin bonded sand mold would have to be contained in a sealed metal box so that when hot metal was poured in and the resin volatilized it could be recaptured. Sodium silicate is another potential binder. This compound can be made on the Moon but it must be dissolved in water then mixed with sand. In the vacuum the water would evaporate, so the job must be done inside a pressurized structure and the sodium silicate must be allowed to dry out before moving the mold outside and pouring metal. Making molds or cores from plaster or cement will also require a pressurized structure where water vapor is condensed from the air. Not only that, there is no way to mix plaster or cement with water out in the vacuum. The water will evaporate in a flash. At least the actual metal casting can be done outside. Casting large parts inside some kind of habitat module would release a lot of heat and that would demand a powerful air conditioning system. It might get so hot inside that only robots can work within.
It seems doubtful that metal objects amassing several tons could be cast inside large inflatable structures even with a concrete floor. Such a structure will still be needed for making sand molds bound with sodium silicate solution and for making molds and cores out of plaster or cement. Airlocks and wheeled carts or pallet jacks will be needed to get large molds outside and to get solidified castings inside.
Rolling mills will require heavy frames to support the rollers. Crude, very crude, castings could be made in cavities dug in the ground. Basaltic mare regolith (m.p. 1150 to 1350 C.) in contact with molten steel at 1500 C. will melt and form a crust on the parts. This crust could be chipped off, ground and wire brushed off by robots. CNC machines could grind the crude frame casting into a finished product. Casting in anorthositic highland regolith which melts at about 1500 C. might be more effective .
8. Grinding Metals
Rough castings will have to be finished by grinding and polishing with CNC milling machines. It is not likely that large heavy parts like steel rollers and extruder barrels would be made entirely by CNC machining of big blocks of steel. This would wear out expensive imported tungsten carbide cutting tools. If it is possible to cast and grind out massive, unitary and simple parts on the Moon and import the complex, lightweight and electronic parts it should be possible to achieve substantial savings.
There are CNC milling machines that can machine 60,000 kilogram (132,000 pound) workpieces, but they are as big as a two car garage  On the Moon, workpieces to be shaped into rollers would only amass about 9,000 to 14,000 kg. (20,000 to 30,000 pounds). In low gravity they would weigh one sixth as much and be easier to move around. There are CNC machines that could work pieces this large and they amass about 16,400 kg. (36,000 pounds). It would seem to make more sense for Moon miners to send up a 16 ton CNC machine that can be used to make numerous machines from rough castings rather than send up complete 40,000 kg. rolling mills and other heavy machines that cannot replicate themselves.
Fig. 8) Hurco VMX84i-50t weight 16,700 kg. (36,806 lbs.) Travel 213 cm X 86 cm X 76 cm (84x34x30 inches). Table size 218 cm X 86 cm (86x34 inches). Work piece weight 5000 lbs. In low lunar gravity a piece that weighs 30,000 lbs. on Earth could be handled. This machine comes close to what is needed on the Moon to make parts for heavy equipment .
Machining requires lubricant. Sometimes water is used. This won't work in the vacuum. It seems CNC machining will have to be done inside the same pressurized structures where molds and cores are made. Tools will get hot but there will not be nearly as much heat released by machining as there would by melting and casting parts weighing several tons. Lubricant/coolant will be recaptured, filtered, cooled and reused over and over again. Working inside a pressurized structure will also protect CNC machines from extremes of lunar temperature.
This is a situation where Peter Kokh's MUS/cle strategy can be applied . “MUS” stands for massive, unitary and simple while “cle” stands for complex, lightweight and electronic. If we can just make the large heavy parts of machines on the Moon and import the electronic controls and perhaps the electric motors the cost of sending cargo to the Moon will be reduced and the industrialization of the Moon becomes more practical.
If CNC machines can make large heavy or massive, unitary and simple parts on the Moon from rough metal castings it will reduce the requirement for transporting heavy objects by rocket and amount to huge savings. Since steel casting presents the problem of the Moon's lack of sand, water and clay, unlike Mars where clay exists, and the need for a pressurized foundry even if the sand, water and clay are available, it might be that CNC machining will be the best way to shape steel parts from crude molds of all sizes that would ordinarily be made by sand casting. The 3D printers could make small intricate steel parts needed in limited numbers including gears, molds and dies and the rolling mills and extruders would mass produce large steel and aluminum plates and sheets, beams, rails, pipes, etc. A combination of pressurized structures where molds and cores can be made along with CNC milling machines, 3D printers, machine shops with drill presses, lathes and other machine tools, solar and electric furnaces and accessory equipment in addition to some other imports should make it possible to produce heavy rolling mills, extruders, large engine lathes and forging presses on the Moon. When all this equipment is working it should be possible to make anything out of metal that has to be made of metal.
The best of terrestrial conventional manufacturing techniques will be applied on the Moon even in the age of 3D printing. Casting is important. There could be times when casting is faster and cheaper than 3D printing; however, casting will require a pressurized foundry or a sealed metal container so that liquid metals don't evaporate into the vacuum. Liquids evaporate in a vacuum. This makes thin film physical vapor deposition (PVD) with molten metals in a vacuum possible. Free vacuum will make PVD easy on the Moon but it can complicate casting. Atoms of molten metal will not reach lunar escape velocity but there could be loss of material due to boil off from molds unless they are sealed or the casting job is done indoors.
Small parts made of aluminum and magnesium could be cast in plaster molds inside the foundry. Plaster, calcium sulfate, would be obtained by leaching anorthite with sulfuric acid. While basalt, steel and iron might be cast outside in the vacuum without to much loss of material by evaporation, wetted sand molds would ordinarily be required to cast these metals and that necessitates a pressurized foundry to recover water vapor from the sand molds that steams off into the air. Sand molds require a binder, usually clay; however, clay is formed by hydrological processes and it will not be found on the Moon, but there might be clay on Mars. It might be possible to use a little chemical magic to make synthetic clay. Polymers and sodium silicate should also be considered. Polymers that burn off in the form of CO2 can still be recycled. The CO2 can be extracted from the air in the casting module and be reacted with hydrogen to form polymer precursors like ethylene. Molten metals will emit lots of heat and a powerful cooling system will be required in the foundry in addition to concrete floors and barriers that can stand up to spilled liquid metal. Concrete is a mixture of gravel, sand and cement. Cement can be produced by roasting highland regolith at over 1500 C. to drive off SiO2 and MgO and enrich CaO and Al2O3 components. If a setting time retarder is needed, some CaSO4 can be made by leaching regolith with sulfuric acid. As discussed previously, a pressurized structure where molds and cores can be made with sand, binders, cement and/or plaster and machining can be done with lubricants/coolants that can be recycled is necessary. This structure or foundry could be made of inflatable Kevlar modules that have concrete floors poured within and are covered with regolith for radiation shielding. Cement will demand a lot of water and it can only be mixed and poured inside pressurized modules. As it dries and sets it will release most of its water into the air and condensors could recover water from the air. Water can come from polar ices. It can also be obtained by combining LUNOX with imported hydrogen or hydrogen from solar wind implanted volatiles mining. Since water is 8/9s oxygen this could be worthwhile.
Casting on Earth seems like a fairly straightforward manufacturing process. On the Moon it becomes rather highly involved. Fortunately, the need to cast anything really huge does not exist. Large metal things like plates and I-beams can all be made outside with rolling mills and extruders. Lunar workers could teleoperate robots that load billets of metals into machines that extrude beams for vehicle frames and weld them up outside with arc welders. Friction stir welding could be used with aluminum. A rapidly rotating ceramic tool is guided along the joint where two metal pieces meet. Friction with the spinning tool generates heat that fuses the aluminum. Much can be made with flat and curved metal plates produced by feeding ingots or slabs of metal into rolling mills. Those plates can be square or workers can laser cut them into various shapes including disks. Beams and rails can be rolled. Beams of various dimensions, rods, bars, rails, pipes, and metal fibers can also be produced by extrusion.
This is the "Lego set" lunar makers have to work with. Rods can make axles. Beams can make vehicle frames. Pipes or tubes can also make vehicle frames. Flat plates can make buckets and ore bins. Disks can be used for wheels and maybe presses can even stamp out wheels. Rails by extrusion are rails. It shouldn't be to hard to make ore cars, rails and buckets and cables for slushers with rolling mills, extruders, presses and a small foundry with machine shop along with 3D printers. A big engine lathe instead of a giant press could spin metal domes from circular metal plates, disks, outside. Beams and rods can make power cable towers and supports for reflector systems. If one is imaginative enough, it might be possible to extrude basalt. Take a billet of basalt, get it red hot and soft, and squeeze out beams for making things like towers and supports. Drill holes in the beams with lasers and bolt them together with steel bolts and one can come up with all sorts of structures. Solar furnaces will require lots of frame members to support reflectors and crucibles. Parabolic solar trough reflectors can be made by rolling and curving sheet metal and dish reflectors can be made by spinning. Hemispheres made by spinning could be welded together to make spherical storage tanks for water, LOX, LH2 and other liquified gases. This work can all be done outside with machines mounted on solid basalt pads with parasols to shield everything from the blistering hot lunar Sun and trap warmth by night.
Forging metals will also be important when this can make parts faster and in larger quantity than 3D printing. Drop forges would have to be very tall and have massive weights in low lunar gravity. Compressed oxygen could drive forging hammers too. Electromagnetic systems are also possible. Since the Moon lacks oil and leakage into the vacuum is likely forging presses would probably not be hydraulic. All sorts of parts can be made from hot metal blanks. All varieties of steel dies might be made by 3D printing; however, printed parts are sometimes more porous and weaker than cast parts. Casting steel dies in the lunar foundry might be called for. The steel dies would be heated and water quenched to harden and temper them. Steam from the quenching of hot metals would be condensed from the air in the pressurized foundry module. Forgings will be in demand. A jet liner contains thousands of forged parts. Rockets, ground vehicles, robots, rovers, refrigeration devices, machine tools and many other things will contain forged parts.
It is true that 3D printing can make some large parts like an airplane wing, but it is slow. It wouldn't make sense to print an I-beam, which would probably be the biggest single part made of metal on the Moon besides parts for heavy equipment. An I-beam would be rolled or extruded outside. There might not be any demand for large I-beams anyway until lava tubes are sealed and pressurized and buildings are constructed within using conventional techniques. Curved plates and domes for "sausage" shaped habitat modules can be cranked out by rolling and spinning. An airlock hatch seems like something that would be cast. It could be possible to stamp or forge hatches outside if there is a big enough press and some disk shaped billets.
Lunar makers must strive to make vehicles and machines on the Moon using 3D printing, rolling, extruding, stamping, forging and only rarely casting. They will have pressurized machine shops where parts are drilled and milled with great precision. There will be assembly shops or garage modules where vehicles and machines are put together. Operating machine tools and assembling things will not generate the extensive heat that melting and casting metals will.
Fasteners will be necessary. Bolts and screws are made in a bolt rolling machine that rolls rods (made by extrusion) between two dies. Nuts and rivets are also going to be required.
To continue with the payload list for setting up a bootstrapping lunar industrial base:
- Inflatable module for foundry with powerful cooling system.
- cement mixer to make concrete for foundry floor and barriers
- electric arc and friction stir welders
- 3D printers as needed
- more solar panels and associated hardware, wiring, etc. to power machines
- very large engine lathe
- cutting lasers, perhaps a cutting table
- forging hammers
- sulfuric acid leaching systems (these might be made of acid resistant basalt on the Moon)
- machine tools (drill presses, lathes, grinders, boring and milling machines, etc.)
- CNC milling machines
- bolt rolling machines
- spare parts for machines--could be printed up on the Moon as needed
There will be basalt poles to support power lines and telecommunications cables. There will be basalt supports for solar panels. Solar farms will cover thousands of square meters of land. There will be melted or sintered regolith roads and railways consisting of steel rails and steel ties on beds of gravel.
Shelter is of utmost importance. It seems that it will be desirable to construct shelter on the Moon instead of continually importing inflatable Kevlar modules from Earth. Aluminum cylinders with domed ends could be used. It might even be possible to make box shaped habitat by welding up flat metal plates and welding in internal supports like webs.
Contour crafting is interesting. This is basically 3D printing on a large scale using cement. However, hydraulic cement won't work outside in the vacuum. The water will evaporate before the concrete can set. This kind of cement could be used with bricks, cement board and slabs to make steps, walls, even furnishings inside of pressurized modules. In sealed and pressurized lava tubes entire buildings could be constructed using conventional techniques. Water would be needed to make all this cement and that would come from polar ices, solar wind implanted volatiles mining for hydrogen and/or imported hydrogen combined with LUNOX as mentioned above. Sulfur cement might be used inside pressurized spaces. Outside on the Moon it might also be possible to use sulfur cement. There are substantial traces of sulfur in the regolith that can be extracted by roasting regolith at up to 1200 C. The sulfur would simply be mixed with sand (sieved regolith) and gravel, heated to 140 C. to melt the sulfur and then poured. A contour crafting gantry could be used to “print up” all sorts of structures. The only drawback is that the extreme heat of lunar day (127 C. at the equator) could melt the sulfur (m.p. 115 C.) and the construction would collapse. Bulk regolith is an excellent thermal insulator. The temperature at a depth of one meter at the equator is a constant -20 C.  If the work is done behind or beneath foil shields and the structure is covered with several meters of insulating loose regolith then it should be safe.
Another possibility is the construction of fused rock structures. The contour crafting gantry would need a bucket consisting of a high temperature resistant metal like molybdenum or tungsten in which regolith was melted and oozed out to form domes of rock layer by layer with layers a few centimeters thick. It might also be possible to use a nickel-steel bucket with an active cooling system. Fused rock structures should be very strong and the material needed to make them is just sitting on the Moon waiting to be excavated. A contour crafting gantry seems like a heavy cargo to the Moon. It seems likely that Moon miners will have the steel beams and other parts needed to build contour crafting gantries plural for doing all kinds of work on the Moon. Two more additions to the lunar industrial seed would be:
- Contour crafting gantries, or just the components that cannot be made on the Moon
- Crucibles made of molybdenum or tungsten with heating elements
Fig. 9 Cross sectional diagram of machine printing cylindrical habitat with extruded molten basalt.
At first, equipment from Earth will be complete with motors. There will be electric motors galore on the Moon in a range of sizes. They will be used for vehicles, digging machines, contour crafting gantries, slushers, railways, rolling mills, extruders, forging hammers, compressors, refrigeration equipment, ventilation fans, water pumps, sewage pumps, 3D printers, small electronics, spacesuit backpack mechanisms, solar panel and antenna tracking systems, coolant pumps, appliances, sewing machines and looms, power tools, machine tools, farm equipment, even electric razors and toothbrushes, etc. A substantial part of lunar industry will involve electric motor production. Aluminum wire will probably be used since copper is so rare on the Moon. Sparse carbon and hydrogen will be combined with plentiful silicon and oxygen to make silicone insulation for the motor windings. Iron and steel will be available. Motors might run on DC or direct current from solar panels might be inverted to 3 phase AC to run 3 phase motors. Such motors can be lighter and more powerful. Entrepreneurs who set up a factory on the Moon that makes a wide variety of electric motors and replacement parts from lunar sourced materials will have a reliable market with a real future. Another payload for the lunar “seed:”
- Electric motor winding machines and any specialized tools needed to get electric motor production going on the Moon
13. Solar Panels
Lunar industry will have a voracious appetite for electrical energy. It would seem reasonable then that solar panels and related gear should be made on the Moon. Dr. Peter Schubert has designed a device that works in a manner similar to a mass spectrometer. The device can produce oxygen, silicon, silicon doped with phosphorus, aluminum and iron; possibly other elements too . This device does not require imported chemical reagents like chlorine and fluorine. It consists of some exotic materials that would have to be imported like thorium oxide and platinum-rhodium. Much of the machine, like the piping systems, waste heat radiators and cryochillers could be made on the Moon from steel and ceramic materials. Boron for p-type silicon is rare on the Moon. Aluminum could be used instead. Phosphorus for n-type material is available. Aluminum could also be used for backing and wiring. Glass could be produced for anti-reflection coatings. More lunar industrial seed cargoes could include:
- Complete and partial Lunar Dust Roasters and All Isotope Separaters
If for some reason this device fails to deliver, then silicon might be produced by serial magma electrolysis. This could be zone refined to a high degree of purity. If this fails, then silicon extraction will require the use of corrosive fluorine imported from Earth. Fluorine could be shipped easily in salt form and electrolyzed to free up the halogen. This will increase costs; however, the cost of imported fluorine will not be as great as the cost of imported solar panels given the huge need for them.
It will not be cost effective to keep importing heavy production machines, robots, bulldozers, habitat, etc. These things must be made on the Moon in greater and greater numbers as lunar industry and populations grow. The Moon cannot “cut the cord” with Mother Earth completely. Computer chips will come from Earth. Factories with clean rooms where integrated circuit chips are made cost billions of dollars. In the more distant future such factories might emerge on the Moon. Until then the mechanical parts for robots, vehicles and other machines will be made on the Moon while the electronic brains are imported.
The various manufacturing methods like 3D printing, rolling, extruding, forging, machining by hand and with CNC (computerized numerical controlled) machines and casting will be used. To replicate some machines large part casting might be required. An inflatable made of Kevlar with more tensile strength than steel with a concrete floor could do small casting jobs, but large parts will have to be cast outside in polymer or sodium silicate bound sand molds. Some carbon and water would not be sacrificed if the mold is sealed. Loss of water or organic chemicals can be avoided if lava tubes of reasonable size can be found, sealed and pressurized with oxygen. Within a lava tube it would be safe to pour hundreds of pounds of molten metals and recover water vapor from the sand molds with dehumidifiers. Living quarters, offices and workshops could be bricked up inside of lava tubes to house hundreds, perhaps thousands, of humans someday.
15. Hypothetical Materials
Ferrosilicon from magma electrolysis might serve as a low performance rocket fuel after powdering. Iron and silicon burn furiously in pure oxygen. Monopropellant might be made from a mixture of ferrosilicon and liquid oxygen. Bipropellant rockets using LOX and FeSi mixed with a carrier liquid, perhaps silane, might be more reliable and safer. Ferrosilicon might not be as powerful a fuel as aluminum but it will be much easier to produce. It can be acquired by magma electrolysis while aluminum production requires imported chemicals and numerous steps. Monopropellants consisting of aluminum powder and LOX have been studied [35 ]. Ferrosilicon deserves more investigation.
Basalt fibers bound with polymer resins are now being used to reinforce concrete. Resins will not be plentiful on the Moon. There will be some carbon for steel and some for agriculture and some for organic chemicals, but those organics will be pricy. Glass fiber reinforced glass matrix composites have been suggested. Unfortunately very little work has been done with this material. The pure silica fibers would add tensile strength and fracture resistance to a matrix of glass that has been doped with sodium and calcium to lower its melting point . Why not a basalt fiber reinforced basalt matrix composite? The matrix could have its melting point reduced so as not to melt the fibers by doping with sodium, potassium, calcium and/or magnesium. It seems that this material would be "easier" to produce than glass fiber reinforced glass composites. Basalt is readily mined and melted. Moon miners have to look at lunar basalt compositions to. Lunar basalt has more iron in it than terrestrial basalt. It might be desirable to alter iron contents with magnetic separations. Casting basalt and fiber drawing in the vacuum and low gravity must be studied. On Earth, makers of basalt products cool the fibers with a water spray. In space, a sealed chamber and a spray of cooled helium gas that is recycled might be called for to cool the fibers.
If basalt fiber reinforced basalt matrix composites are feasible and cheap, then this could be a material for space frames for power satellites, space stations, space shipyards, large telecommunications platforms and space telescopes. No imported reagents would be needed. Manufacturing this stuff, which would have a density of about 2.95 vs 2.7 for aluminum, in quantities needed for solar power satellite construction should be cheaper then producing aluminum.
The primary structure of a powersat is about ten percent of its total mass or about 5,000 tons for a 50,000 ton 5GWe SPS and 10,000 tons for a 100,000 ton 10 GWe satellite  If 20 powersats a year are built and each is rated at 10 GWe then in 50 years there would be 1000 of them and 10 TW of power. That's a vast amount of energy. At present the world uses about 15 TW and it is predicted that civilization will demand about 30 TW by 2050 A.D . However, two-thirds of this is waste heat, thus 10 TW of electrical power could energize the world if everyone switches to electric heat and electric cars About 200,000 tons of aluminum would be needed annually just for SPS frames. This would require quite a bit of infrastructure on the Moon. It would require the costly importation of halogens to make electrolyte using various proposed methods for electrowinning aluminum from lunar regolith. Fiberglass or basalt might be cheaper. If basalt fiber reinforced basalt composites are possible and cheap, this would drastically increase the value of lunar basalt and make the case for a mare/highland coast installation much stronger. Basalt composites reinforced with silica fibers or fibers made from melted and drawn anorthositic highland regolith should also be investigated.
Insignificant traces of copper exist in regolith. Calcium conductors might serve as an alternative to aluminum. According to Geoffrey A. Landiss,”An alternative material for electrical conduction is metallic calcium. This is not used on the Earth because of the extremely high reactivity of calcium in an oxygen atmosphere, but this may not be a problem on the moon.
Calcium has the property of having one of the highest conductivity to density ratios of any easily-available metal, that is, metallic calcium makes the lightest wires. However, aluminum is adequate for wire use, and aluminum technology is well developed...”
Solar power satellites could benefit from lighter conductors. Mass driver coils might be made of calcium too. That could be of value for mass driver propelled spaceships. Calcium might be extracted from slag produced during extraction of other metals. This warrants more investigation.
16. The Bottom Line
It could be true that history is governed by economic factors. If so, then lunar development will be governed by costs and profit margins, especially if the Moon is developed by private entrepreneurs who have to keep an eye on the bottom line. It seems reasonable that products that can be produced on the Moon with lunar available resources alone will be cheaper than products that rely on imports, but reality and the marketplace can defy common sense. For instance, it would seem that it would be cheaper to pump up oil from deep wells in the USA rather than ship oil in supertankers across thousands of miles of ocean, but it's not. Presently, there is no way to predict any of the costs of doing business in outer space. Rocket launches cost tens of millions of dollars just to reach low Earth orbit and they explode frequently. Reliability must be improved. Prices for rocket launches to LEO will probably come down in the future. That seems to be the trend for so many products be they aluminum, computers or automobiles and microwave ovens. Even if the price for a rocket launch comes down by a factor of ten to one hundred, it will still be expensive to travel in space and the use of on site materials and energy will still be preferred.
1. G.K. O'Neill, The High Frontier, Human Colonies in Space, William Morrow and Company, Inc., New York, pp.. 135-136, 1976.
2. David R. Criswell, The Initial Lunar Supply Base, http://www.nss.org/settlement/nasa/spaceres/V-1.html
3. R.A.Freitas, Jr. and W.P.Gilbreath (eds.), Advanced Automation for Space Missions, NASA Conference Publication 2255, National Aeronautics and Space Administration Scientific and Technical Information Branch, 1982. https://en.wikisource.org/wiki/Advanced_Automation_for_Space_Missions,
4. SpaceX website <available http://www.spacex.com/falcon-heavy
6. Marcus Lindroos, Lunar Base Studies in the 1990s 1993: LUNOX, http://www.nss.org/settlement/moon/LUNOX.html
8.Peter Kokh, Lunar Prospector and the Lunar Frontier, 2007, < http://www.asi.org/adb/02/lunar-frontier.html >
9. Wolverton Environmental Services, Inc., Indoor Air Pollution, < http://www.wolvertonenvironmental.com/air.htm >
10. Al Globus and Joe Strout, Orbital Space Settlement Radiation Shielding, preprint, pg. 14, 2016. < http://space.alglobus.net/papers/RadiationPaper.pdf >
11. W. David Carrier III et. al., Lunar Sourcebook, Chp. 9, “Physical Properties of the Lunar Surface,” pg. 484, < http://www.lpi.usra.edu/publications/books/lunar_sourcebook/pdf/Chapter09.pdf >
12. Richard E. Gertsch, A Baseline Lunar Mine, NASA SP-509, vol. 3, 1992 < http://www.nss.org/settlement/nasa/spaceresvol3/ablm1.htm >
13. Larry A. Haskin, Hydrogen and Oxygen From Lunar Polar Water, Or From Lunar Dirt? Magma Electrolysis, http://meteorites.wustl.edu/abstracts/nvm3_00/a_m00h02.pdf
14. Dr. William Agosto. "Lunar Beneficiation." http://www.nss.org/settlement/nasa/spaceresvol3/lunarben1b.htm.
15. Rudolf Keller and D.B.Stofesky. "Selective Evaporation of Lunar Oxide Components." Space Manufacturing 10, pp. 130-135, Abstract: http://ssi.org/ssi-conference-abstracts/space-manufacturing-10/
19. Advanced Automation for Space Missions. Chp. 4.2.2 http://en.wikisource.org/wiki/Advanced_Automation_for_Space_Missions/Chapter_4.2.2
20. Lunarpedia, Volatiles, http://www.lunarpedia.org/index.php?title=Volatiles
21. Handbook of Lunar Materials, NASA Reference Publication 1057, 1980, pg. 115, http://www.nss.org/settlement/moon/library/1980-HandbookOfLunarMaterials.pdf
22. Matthew E. Gajda et al. "A Lunar Volatiles Miner." nasa-academy.org/soffen/travelgrant/gadja.pdf
23. Rudolf Keller and D.B.Stofesky. "Selective Evaporation of Lunar Oxide Components." Space Manufacturing 10, pp. 130-135, Abstract: http://ssi.org/ssi-conference-abstracts/space-manufacturing-10/
24. Christian W. Knudsen and Michael A. Gibson, Processing Lunar Soils for Oxygen and Other Materials, http://www.nss.org/settlement/nasa/spaceresvol3/plsoom1.htm
25. Willam N. Agosto, Lunar Beneficiation, http://www.nss.org/settlement/nasa/spaceresvol3/lunarben1.htm
26. Lawrence A. Taylor and W. Carrier III, Oxygen Production on the Moon: An Overview and Evaluation, Resources of Near Earth Space, 1993, http://www.uapress.arizona.edu/onlinebks/ResourcesNearEarthSpace/resources04.pdf
27. Larry A. Haskin, TOWARD A SPARTAN SCENARIO FOR USE OF LUNAR MATERIALS , Lunar Bases and Space Activities of the 21st Century, W. W. Mendell, Editor, 1985, Lunar and Planetary Institute, http://www.lpi.usra.edu/publications/books/lunar_bases/LSBchapter07.pdf#pagemode=bookmarks&page=3
28. W. Howard Poisl and Brian D. Fabes, Refractory Materials from Lunar Resources, pg. 355, http://www.uapress.arizona.edu/onlinebks/ResourcesNearEarthSpace/resources14.pdf
31. Peter Kokh, "MUS/cle Strategy for Lunar Industrial Diversification," Lunar Reclamation Society, 1988 < http://www2.moonsociety.org/publications/mmm_papers/muscle_paper.htm
32. David Schrunk et. al. The Moon: Resources, Future Development and Settlement. 2nd ed. Praxis Publishing Ltd. Chichester, UK, pg. 54, 2008.
33. Dr. Peter Schubert. "How Moonrocks can Save the Earth." 2008 < http://moonsociety.org/spreadtheword/pdf/UsingMoonRockstoSavetheEarth.pdf >
34. Geoffrey A. Landiss , “Materials Refinning for Soalar Array Production on the Moon,” NASA/TM—2005-214014 <https://info.aiaa.org/tac/SMG/SCTC/Shared%20Documents/Landis-Lunar_Production-TM-2005-214014.pdf >
35. Wickman Spacecraft and Propulsion Co,, “Rocket Propellant from Lunar Soil,” http://www.wickmanspacecraft.com/lsp.html
36. Peter Kokh, “Spinining Up Glass-Glass Composites Technology,” Lunar Reclamation Society, 1987, < http://www.moonsociety.org/publications/mmm_papers/glass_composites_paper.htm >
37. Solar Power Satellite System Definition Study, Vol 5, Phase 1, Executive Summary, Final Briefing, Boeing, pg. 113, 1978. < http://www.nss.org/settlement/ssp/library/BoeingCR160374-1978DEC-Phase1Vol5FinalBriefing-ExecutiveSummary.pdf >
39. Geoffrey A. Landiss, Aluminum Production on the Moon with Existing Silicon Production, http://www.asi.org/adb/02/13/02/aluminum-production.html