Helium

The Moon is an abundant source of He3. He3 has a high market value, even though He3 fusion has not yet demonstrated net energy production (break even). It might be worth collecting He3 from the Moon today simply to sell into the existing terrestrial market.

Current market price for He3 is about $46,500 per troy ounce ($1500/gram, $1.5M/kg), more than 120 times the value of gold and over eight times the value of Rhodium.

Question: can we reduce the cost of recovering He3 from the lunar surface to that level, e.g. $1500 per gram? What would be the capital cost of setting up a small He3 production facility on Luna?

Would it depress the market price today? That depends on the size of the market, and there is not much data on that.

The US tritium and helium-3 stockpile sizes are classified, because they give a hint as to how many US nuclear weapons are still functional. According to Wikipedia “approximately 150 kilograms of it (He3) have resulted from decay of US tritium production since 1955.” We could assume a similar quantity has been accumulated in the ex-USSR, and perhaps additionally from other thermonuclear powers (UK, France, China).

Today, the world's supply of Helium-3 can probably be counted in hundreds of kilograms, value of 100 kg would be $150M. So the total stockpile value today is probably about half a billion USD.

The US DOE does sell He3 commercially, but how much of the present stockpile has actually been sold on the open market? Not sure if that number is publicly available.

But for arguments sake let us start at the level of collecting 100kg of He3 from the Moon and assume its value would be $150M.

Well alas even those number do not look good.

The cost of soft landing even a small probe on to the lunar surface would easily cost that much or more. How much He3 could a small lander manufacture? How many grams per day?

Well that of course depends on the production method.

A commonly discussed method is cooking the regolith to about 1400 degF or 760 deg C. This would require a lot of energy, requiring the lander to have either a nuclear source, or large solar panels.

Basalt has specific heat capacity of 0.24 cal/g/degC or 0.84 KJ/kg degK.

To heat 1kg of basalt by 700degC requires about 600 KJ

Best lunar regolith (Maria) is 0.01 ppm of He3

So the 600 KJ will yield 0.01 milligrams of He3

So 600 Watts power source could produce 0.01 mg He3 per second = 0.6 mg/minute = 36mg/hour = 864mg/day = 315 grams per year

Whether this business concept is viable depends on how quickly we want to amortize our investment.

Let us say our target is to produce 100 kg He3 in one year, then we need a power source of about 200 KW. That would give us a revenue stream of $150M per year assuming the He3 market does not become flooded and the price drops.

How much would it cost to set up a 200KW power source on the Moon?

A solar based system would be in darkness 50% of the time, so would need to operate at 400 KW. If it were on a lunar polar mountain top it might be in near continuous illumination. Let us assume that best case of 100% lighting.

Assuming PV10% efficiency and a fully steerable array, this would need an area of about 2,000 square meters, or about 45 metres square.

A simple non-PV solar reflector could be near 100% efficient, needing only 200 sq-m or 14 metres square, or aperture.

Setting up a 14 m aperture mirror on the Moon would be a major engineering challenge, although fortunately would not need to be particularly accurate, certainly nothing like as difficult as an astronomical telescope mirror.

How much would it weigh?

Would a nuclear power plant have better performance per kilogram of lander payload?

Maybe other contributors are interested to develop these lines of thinking.

More thermal analysis needs to be done. For example, might it be possible to recycle the heat using some form of cogeneration. Such as use the hot waste regolith, after it has been processed, to pre-heat the next incoming batch of raw dust, and thus reduce the number of solar joules needed?

That could greatly reduce the size of solar array needed and/or significantly increase the system mass throughput.

Extraction

Baked Regolith

A commonly discussed method is cooking the regolith to about 1400 degF or 760 deg C.

Here is a reference with some details (1989, H. H. Schmitt et al):

http://fti.neep.wisc.edu/pdf/fdm817.pdf

They describe three steps: 1) heat to a few hundred deg C to drive off the volatiles 2) fractional distillation to decant off the heavy volatiles 3) separate He3 from the He4 using standard superleak process

Two challenges are devising a method to process large quantities of regolith as the He3 is at a low concentration, and providing a high power thermally efficient heat source on the Moon.

This would require a lot of energy, requiring the lander to have either a nuclear source, or large solar panels.

Basalt has specific heat capacity of 0.24 cal/g/degC or 0.84 KJ/kg degK.

To heat 1kg of basalt by 700degC requires about 600 KJ

Best lunar regolith (Maria) might be 0.01 ppm of He3

So the 600 KJ will yield 0.01 milligrams of He3

So 600 Watts power source could produce 0.01 mg He3 per second = 0.6 mg/minute = 36mg/hour = 864mg/day = 315 grams per year