The pieces for large scale nuclear fission on the moon.
1. Hyperion Power Generation Uranium Hydride reactor maybe ready 2013.
– if fusion is available because of success with IEC fusion or colliding beam then those could be used instead
– If lightweight molten salt reactors were developed those would be better fission reactors
2. Moon mining and extraction of uranium (tech available now just cost and effort)
3. Laser Enrichment ready 2012
4. High efficiency thermoelectrics 2015-2017 (optional for weight saving)
5. A reasonable heavy lift rocket system to deliver 20 ton pieces to the moon
6. As reader Will Brown notes in the comments, lunar cement can be made easily. the lunar cement recipe is for about 600 kilograms (1300 pounds) of Moon dust, 60 kg (130 pounds) of epoxy, 6 kg (13 pounds) of carbon nanotubes and less than a gram of aluminium. This would reduce the weight of cement for the moon by ten times.
7. Initially the mining and enrichment industry for the moon can be avoided by taking a sufficiently safe form of uranium hydride fuel which requires little processing.
The first of the compact fission reactors will not be ready until 2013 at the earliest and could be delayed a few years.
Mining lunar Uranium (even in the its low concentration is possible).
GE Silex Laser enrichment should have its first commercial facility operating in 2012. This would be the better way to enrich the uranium.
By 2020, there will likely be very efficient thermoelectric material for converting heat to electricity so that a heavier steam generator would not be needed.
The Hyperion Power Generation uranium hydride liquid metal fission reactor will weigh fifteen to 20 tons, depending on whether you’re measuring just the reactor itself or the cask—the container that we ship it in—as well. It was specifically designed to fit on the back of a flatbed truck because most of our customers are not going to have rail. It’s about a meter-and-a-half across and about 2 meters tall. It will generate 27-30 Megawatts of electrical power from 70 MW of thermal power. This means 0.5 to 0.75 tons per MWe for the nuclear reactor. The steam turbine to convert the power is counted separately. The core size is about one seventh the size of compact sub reactor cores.
Mining Lunar Uranium
Launching a loaded uranium reactor to the moon would not be possible because of the public relations issues. The unloaded reactors could be sent with heavy lift capability. Then a compact system to process the concentrations of Uranium and thorium on the moon would have to be built on the moon. There are uranium concentrations on par with uranium in granite on earth in the better locations on the moon.
The KREEP will consist of complex minerals made mostly of silicon, oxygen, aluminum, iron and magnesium with some other good stuff like phosphorus, potasssium, uranium and rare earth metals like lanthanum (used to increase the refractive index of glass) and hafnium.
we must do is break down the crystal matrix like we do with other minerals by melting, quenching and grinding. Iron bearing material can be removed with magnets. Then we can carbochlorinate the stuff by mixing it with carbon dust, exposing it to a stream of chlorine gas (both C and Cl will be carefully recycled and replenished by volatile mining) and heating it with solar reflectors or lenses. This will convert the stuff to chloride salts like that which we find in seawater. The silicon tetrachloride will boil off at only 56.9 C. It will be decomposed with solar heat to get pure silicon for solar panels and recover chlorine gas. Aluminum chloride will sublimate at 178 C. It can be recovered and electrolyzed to get aluminum. Carbon monoxide will also form and vaporize off to be recycled by reaction with hydrogen for conversion to methane and water which can be pyrolized and electrolyzed respectively to recover hydrogen, carbon and get some oxygen. The chloride salts that remain will be dissolved in water and pumped through plastic filters to get the uranium.
We can imagine other plastic filters that will absorb phosphorus, potassium, rare earths, thorium and other trace metals perhaps. After uranium filtration, the salt laden water will be boiled down, condensed, and the metallic chloride salts will be decomposed with extreme solar heat in a ceramic retort to recover chlorine, or they will be subjected to electrolysis. Note that the nuclear scientists also filtered titanium, vanadium and cobalt out of seawater, so we should capture these in the process of filtering out uranium if they are present.
Separating those metals from each other could be problematical. Uranium can be fluorinated to make UF6 which has a low boiling point, so we could just do that to roast it out of the mix of metals we filter out of our salt solution.
There are methods to get uranium from such low concentrations but tend to be expensive. However, even $1000-10000 per kg to recover low concentration uranium from the moon would be worthwhile.
Natural uranium in our solar system is about 0.7% U235 and the rest U238. The Hyperion Power Generation reactor uses 4.9% U235.
If we were to use the General electric/silex laser enrichment (being deployed commercially in 2012) that could then bring the material up to sufficient enrichment efficiently and more compactly.
Estimate the silex laser process needs 5-50 kWh/SWU [separative work units], actual details are classified but based on the 2-20 times better gas centrifuge statements.
A kilogram of LEU requires roughly 11 kilograms U as feedstock for the enrichment process and about 7 separative work units (SWUs) of enrichment services. To produce one kilogram of uranium enriched to 3.5% U-235 requires 4.3 SWU if the plant is operated at a tails assay 0.30%, or 4.8 SWU if the tails assay is 0.25% (thereby requiring only 7.0 kg instead of 7.8 kg of natural U feed).
How Efficient Will the Hyperion Power Generator Be?
Based on a re-examination of the wording of the uranium hydride reactor patent by Otis Peterson, the reactor can burn 50% of fissile material (the odd number isotopes and not Uranium 238). Thus the efficiency of fuel burning is probably about 67% better than current reactors. Current reactors burn 30% of the starting fissile material. This would suggest being in the range of 60-90 gwd/t (gigawatt days per ton.)
Hyperion’s CEO has stated a football size amount of waste is what is left after 5 years of 70 MWth, which would only be about 140 kg. If it started with 300 kg of uranium.
300 kg of uranium has 300,000 MW days of power if converting 100% of fissionable material (the starting uranium and its fissionable products. (1000 gwd/t * 0.3 tons)
The 70MWth for 5 years would need 5 *365*70 = 127,750 MW days
This would suggest that 42% of the available fissionable power.
This site is confident about advancing nanoscale manufacturing capability. This should lead to nanomaterial thermoelectrics with ZT scoreof 4-20 and possibly higher. This would be as good as steam generators efficiency in a lighter weight system.
Fusion power may not be ready until later. The best bet being the Tri-alpha energy colliding beam fusion and the EMC2 Inertial electrostatic fusion.