Skip to content

Sciencey science stuff

Years after initial space-mining ventures went bust, startup AstroForge has announced two missions in 2023 to obtain rare-earth minerals from a near-Earth asteroid.

Rare earths, from an asteroid? Given that concentrates are worth, hmm, a few thousand $ a tonne, seems like a remarkably expensive method of getting them.

The platinum-group metals (PGMs) — iridium, osmium, palladium, platinum, rhodium, and ruthenium — which are among the rarest mineral commodities in Earth’s crust. Just 30 tonnes of rhodium, used in catalytic converters, are mined every year, and only three tonnes of iridium. Mostly these minerals come from mines in South Africa, Siberia, with some mines in the U.S. and Canada.

That’s not a good start to a sciencey, science discussion now is it? Platinum group metals are not rare earths, rare earths are not platinum group metals. Sigh.

“The appeal of asteroid mining is elements that are rare in the Earth’s crust may be found near the surface of some asteroids, where they could be relatively easy to access,” says Michael Brown (Monash University). “But developing the technology to robotically and effectively mine tons of raw material from distant asteroids won’t be easy.”

AstroForge plans to start small, literally, with its first CubeSat the size of two loaves of bread. Its first mission will test in-situ refining in a zero-gravity environment. “That’s really the piece that we see as the highest risk because it’s unproven technology in space,” says Acain. “So we’re launching a CubeSat up to low Earth orbit to understand and characterize our refinery in those harsh environments.”

Sounds feasible to me. Not that I actually know much about this but yes, iron rich asteroids are likely to have high nickel and so also pgm contents. You can refine them with a lot of ‘leccie – solar cells therefore. Why not?

The long-term plan is to have much larger spacecraft mine the surface of M-type asteroids and return to Earth only refined PGMs. It’s a big jump. “Scaling that up to return commercially viable quantities of processed material from asteroids millions of kilometers away is going to be difficult,” says Brown.

And that’s where the problem is.

“I hope the satellites are successful, but there are good reasons for caution,” said Brown. “The small satellites that will be flown in 2023 have masses of kilograms and budgets of millions, but commercial space mining missions would have masses of many tons and budgets of billions.”

Yes, a distinct problem.

AstroForge plans to take it milestone by milestone, but Acain thinks it’s an absolute necessity to extract these rare minerals off-Earth. “We have a finite supply of resources here on Earth – that’s a fact – and there’s more demand for these resources than ever before,” he says. The present mining process, he adds, is costly and polluting. “Taking it off-Earth is the only way we see to solve all of these issues.”

Maybe. But at what price? After all, we’re talking business here so price is an important consideration. Which brings us back to the opening line here:

Years after initial space-mining ventures went bust

As I wrote back then:

It’s also true that those nickel iron asteroids are likely to be rich in platinum-group metals (PGMs). They too can be refined with a bit of electricity, and they’re sufficiently valuable (say, for platinum, $60m a tonne, just as a number to use among friends) that we might be able to finance everything we’re trying to do by doing so.


Start from the size of the platinum market. This is some 6.2 million ounces a year. 6.5 million ounces of virgin material, that is: given the value of the metal some to all of past usage is recycled as well. At our $2,000 an ounce price guide, that gives us a market value of some $13bn a year. That certainly seems large enough to keep a space programme running. (Do note, I’m ignoring palladium, a similar sized market, and rhodium etc, which are much smaller ones. They don’t change the final conclusion by their inclusion or exclusion.)

Except that’s not quite how markets work. There are demand curves as well as supply ones: sure, a nice high price will encourage new entrants like Planetary into the market. But in order to shift all this new material, prices will have to decline. The important question therefore is how elastic is the market? How far, if at all, will the price fall if a new supplier enters?

From a recent trade report we’ve seen recently, an extra 250,000 ounces has come onto the market. This has led to a 25 per cent fall in the price of platinum. Ah! Price is very sensitive to an increase in supply, then. Or, if you prefer, demand is very insensitive to a change in price. They’re the same statement, really.

Ah. If you start bringing back platinum in sufficient volume to pay for the billions in satellite mining costs then you crash the price against your output of platinum.

Ooops! And the more you mine to cover your costs the lower the price received. For the price is to do with the rarity and if you’ve just gained another whole source then the rarity rather goes away, doesn’t it?

My opinion does not change therefore:

Guys, I wish you all the luck in this little blue marble of ours, but I do think this is best described as an adventure, not a business.

Spotter, VikingWill

34 thoughts on “Sciencey science stuff”

  1. Nice idea and I agree with you mostly but

    “If you start bringing back platinum in sufficient volume to pay for the billions in satellite mining costs then you crash the price against your output of platinum.”

    Isn’t that dependent on how one sells the platinum? Shirley being the sole owner of vast amounts of something means that one can set the price ?

    Also, if the amounts of platinum and palladiun are pure enough and ( big) IF a Fleischman Pons style reactor can be made to work, then it might make such industrial scale technology feasible.

    As so often, it is the use of, rather than the raw material itself that brings in the dosh.

  2. To get really rocket sciencey about this you need to understand the tyranny of the rocket equation

    Getting an empty spacecraft to your asteroid is doable. Getting one back with a payload is exponentially harder as you also need the propellant for the return journey. As it’s not there you must take it with you. You also then need extra propellant to “land” this return propellant on the asteroid, but you need even more propellant to get the return plus the landing propellant from Earth to the asteroid. All of this heavy propellant needs launching in the first place, and launching this heavy propellant into space means even more propellant is needed to launch it. Basically the maths shows it to be an exponential progression.

    The more payload you want to get back, the more propellant you will need to deliver to the asteroid you are mining. You will need exponentially more propellant to do so. It’s not a good business model.

  3. If they can demonstrate the ability to land on an asteroid and return a sample, then that could create other opportunities. For example, selling science missions to collect samples from any low-gravity body in the solar system at a lower cost than a bespoke system built by a government space agency. If they can scale it up to work on larger bodies like the moon, that’s another potential market. They might also make money from licensing their technology, if it’s good enough. But the surest way to make money in a gold rush is to sell shovels to the prospectors, so maybe the path to really serious profit in this business is to do enough to trigger a rush and then sell or license the tools.

  4. @AndyF:

    I suspect if challenged on that they’d start hand waving about solar sails. I’ve no idea whether sails would be feasible for this (too many other feasibility questions to worry about), but at least they don’t need fuel.

  5. I have lot of doubts about this science stuff in the first place. Because I have doubts about the science. When it comes right down to it we don’t know what’s out there until we go out there & look. All the sciency science stuff at the moment is largely built upon assumptions on top of assumptions on top of assumptions. And I’m sure some of those assumptions are assumed because they’re assumptions that produce healthy research grants for researchers.
    It’s like this ongoing search for planets around other stars. Just at the moment it’s restricted by only having a couple of orbital telescopes & what can be done from the ground. So it comes up with a lot of large planets circling close to relatively small stars because they’re the easy ones to detect at the current state of the art. However we continue to pay for the salaries & equipment for hoards of astronomers to find more of them. Why? 50 years time, at the rate we’re going, we’ll be able to put telescopes up either in orbit or on the moon would resolve an earth sized planet in an earthlike orbit a couple hundred light years away. So why not just wait? They ain’t going nowhere. Why do it badly now when you can do it better & cheaper later? Something to do with keeping those astronomers’ wage checks coming?
    So back to the asteroid mining. It makes no difference what metals asteroids are made of. Unless you solve what AndyF talking about it’s irrelevant. Maybe you can solve it by making your fuel out of asteroids but one suspects they’ll the carbonaceous ones not the metallic ones. When you’ve only sampled the surface couple of millimetres of one of them, it’s a bit hard to tell at this stage. Sounds like AstroForge is another iteration of my previous paragraph but fleecing private mugs rather than government ones.
    I suspect the whole of astrophysics is another one. That some time in the future, some rather important assumptions will turn out to be bollox & the universe isn’t the slightest bit what they currently think it is. So goodbye to the ever elusive dark matter & energy they’ve been using to make their sums come out right.

  6. BiS
    “I have lot of doubts about this science stuff in the first place. Because I have doubts about the science. When it comes right down to it we don’t know what’s out there until we go out there & look. All the sciency science stuff at the moment is largely built upon assumptions on top of assumptions on top of assumptions. And I’m sure some of those assumptions are assumed because they’re assumptions that produce healthy research grants for researchers.”

    You’re not the only one with doubts. Here’s a nice video from a septic particle physicist.

  7. I think you missed the point, Otto. The rocket carrying the fuel has to to be several times the size of the one carrying the payload. Depends on the fuel:payload ratio. But that’s very unlikely to be 1:1. It’s a similar problem to getting out of the earth’s gravity well. Requires 4600 tons of fuel to put 150 tons into LEO. That’s 9.4 km/sec delta V. To get from LEO to the asteroids you need another 7 km/sec delta V. You could probably get some of that from gravity assist flybys of assorted solar system objects, but it’d still take a lot of fuel.

  8. And think it through some more. The earth return payload has to be multiples of mining payload to make the project economically viable. You’re going to be talking in $100’s/kilo to be way on the optimistic side & a mass ratio of 1000:1? And all that mass needs 7km/sec delta V to return it to earth orbit. You’re not really talking about one rocket. It’s more like a fleet.

  9. (sighs as joke goes flying over BiS’s head )

    I am reminded of Emperor Franz Josef who when told that it took two engines to get a train to the top of the Simmering, where his holiday home was, but only one to get back down, allegedly remarked ” Well isn’t it rather crowded with all those locomotives up there ?”

  10. Also needed for any of this remote space mining / refining to work are massive advances in robotics and “AI”. You need reliable, adaptable machines that can do a lot of “thinking” on the go.

    At the moment it’s an Underpants Gnome business model.

    [ go to space ] —-> [ ? ] —-> [ profit ]

    That question mark does a lot of work in science fiction, which is fine, but it’ll be a tough sell even for the most adventurous investors.

  11. @ Arthur the cat

    Well given time a solar sail would do just fine. LightSail 2 (not much of a payload) does 0.058mm/s2 so it can increase its speed by 549 km/h in one month of sunlight. The asteroid belt is ~200-600 million kilometers from Earth so its going to take a while to get back (well close enough to be rescued by some sort of tug). Probably tens of years. It would be a long investment and that long knowledge of when the super valuable payload gets home would play merry hell with the futures market.

  12. This is overlapping with the earlier thread, Artifical Intelligence IN SPAAACE!!!!

    For the fuel problem, couldn’t you use the Omian Army method, send out some fuel, station it, send out another fuel ship, it uses the stationed fuel to get further, stations its fuel payload, send out another ship, it uses the previous stations to put another fuel station further out, repeat until you have a string of coaling fuelling stations. With exponential equations, 100 1kg fuelling rockets is likely to be more efficient than one 100kg rocket.

  13. @jgh

    In practice you do something like that. It’s why rockets are two or more stages. The first stage lifts itself and its huge mass of propellent plus the fully fueled second stage and payload. By the time the first stage is empty and it’s time for the second stage to take over, the much smaller second stage is high up and moving fast. At this point the first stage is discarded so the second stage does not have to carry the weight of the empty second stage all the way to orbit. If it tried to do so it would run out of propellant and fall back to earth before reaching orbit.
    Curiously it doesn’t make much efficiency difference between big and huge rockets. It just depends on the the mass ratio of the empty to fully fueled rocket. Ten times the size of rocket equates to ten times the size of payload.

  14. @AndyF
    Sure you could a light sail to get out there. It’s getting back that’s the problem. Can you tack with a light sail? The sunlight is coming from 90° of the direction you want to shed delta V. It’s not a windsail where you can tack into the wind via aerodynamics. It’s solely light pressure.

  15. To add to the above, I think you can but it’s counterintuative. You require delta V in the opposite direction to which the asteroid is orbiting. Less delta V produces a lower solar centric orbit. If the plane of the sail is set at 45° to the incoming sunlight, it both pushes the sail away from the sun but also in the correct direction. Now orbital dynamics. Raising an orbit reduces velocity but gains gravitational potential. The trans orbital delta V element reduces velocity, which cancels out the added gravitational potential. You have to counter the gravitational potential or you end up with an elliptical orbit which crosses the earth’s orbit too fast to capture.I think

  16. Peeps forget something….

    It doesn’t take that much fuel to drop down into a gravity well… All you need to do is brake a bit from orbital velocity..
    A bit more accurate braking if you want to have anything at a given time at a given velocity at a specific place, but still far less fuel than you need to get up to that orbital velocity ( 13-ish km/s for the asteroid belt) to begin with.

    It’s how the Japs got those asteroid samples back to Earth to begin with…

  17. Come on now chaps – surely you’ve read the standard stuff? You don’t send a continuous stream of rockets to and back from your asteroid. That would be silly.

    The standard theory about asteroid mining and the like is that set you set up a base near/on your asteroid and work there. You use solar mirrors as your local power source, and expect to find something you can use as reaction mass. You use local resources (the asteroid and the solar energy) to build nasty crude craft which you will launch earthwards. They may have complex safety-conscious electronics but most of the mass comes from the poor gutted asteroid. And electronics doesn’t mass much, so even if you need to launch stuff outwards from earth you get lotsa electronics per payload.

    Then you build a pipeline of the nasty crude craft sent towards earth. You use the reaction mass – solid or liquid – to slow your craft so they drop into the gravity well. You use the reaction mass to steer as you wish. Small electromagnetic ‘machine guns’ can fire mass as toy wish to steer; don’t need rockets/fuel/gas. You may well have a cunning net thing or other catch ’em and maneuver them sited closer to earth. Then you faff with the goodies in lunar orbit or whatever, and make even more money shipping finished goods to the gaping maw of civilization (you can deliver giant chunks of ore or metal if you like, but folk get edgy with giant things falling out of the sky at them)

    If you have deployable nuclear power then you can use the instead of/as well as the solar mirrors.

    End result is that you don’t have to keep sending rockets out and rockets back.

    As BiS notes, for this to work rather depends on the asteroids being made of the right stuff, but that’s why you go looking first.

  18. What BIT sez, with one addendum..

    People automatically assume a lot of the structural stuff to build an asteroid refinery has to come from Earthside.

    It doesn’t. The design parameters won’t be based on standing up to a 1g vertical gravity field, but on what’s needed to pick up the forces locally involved in a microgravity field.
    You can build up from a pilot plant using the slag from the refining process plus whatever rock you have lying around.
    Weight is not a problem. Mass may be if you want to move around a lot, resultant size for a given strength may be, but there’s no weight problem when building/designing stuff. You can get chunky except for the places where things are pretty specific/demanding.

    This means that you can actually make a fair amount of what you need onsite from whatever is not interesting enough to drop down, but priceless where you’re operating. For “Free”.

  19. And don’t forget that thanks to Musk we can use the same rocket multiple times now, drastically reducing the cost of getting stuff up there..

    The SLS is impressive, but it is Stupid beyond anything I can politely think of…

  20. It doesn’t take that much fuel to drop down into a gravity well… All you need to do is brake a bit from orbital velocity..
    A bit more accurate braking if you want to have anything at a given time at a given velocity at a specific place, but still far less fuel than you need to get up to that orbital velocity ( 13-ish km/s for the asteroid belt) to begin with.

    I think you need to understand orbital dynamics, Any orbit can be expressed in delta V. The acceleration needed to achieve that orbit. Acceleration has a sign. + or – relative to the starting condition. To get from circular earth orbit to circular asteroid orbit requires delta V+. At asteroid orbit the spaceship will have acquired gravitational potential. The same as you do when a lift takes you to the top of a tall building. What you’re saying is that you can step off the roof directly on to the ground without the intervening plummet. (Gravitational potential.) The delta V- To get from asteroid orbit to earth orbit takes the same delta V as in the other direction. Orbit’s not a place, it’s a velocity & vector.

  21. It’s how the Japs got those asteroid samples back to Earth to begin with…

    Not really, 25143 Itokawa and 162173 Ryugu are both (potentially hazardous) Apollo asteroids with orbits that already intersect Earth’s. So the fuel used by the probes to reach their orbits also accounted for most of their journey back.

    Moving from, say, an orbit in the asteroid belt to Earth’s orbit is a vastly different energy equation.

  22. You can’t build your plant in space.

    Your plant needs to be built out of purified and worked material. Which means sending a smelter and processing plant first.

    This is not a bootstrap opportunity. You can’t start working on your plant until you have a plant to build the plant you want. At which point, why not just send your original plant?

    At some point, if someone set up in space a plant that could mine, sort, smelt, allow and then work the material, then it would be become workable. But that would cost the world’s GDP at the current cost of getting the material into space.

    I fancy that dealing with extremely hot molten metal in zero gravity and with limited methods to dissipate the heat would create an environment that would be, um, challenging.

    It ain’t happening this century.

  23. Must agree that it’d probably not be economic Chester. But think of the European age of expansion. Was the Portuguese empire economic? The initial Spanish push to the west wasn’t. Until they found all those Aztecs and Incas to loot.

    Certainly the British colonisation of Oz was uneconomic. Or lets look at the late 19th century Scramble for Africa.

    I suspect that for one ratbag reason or another, someone will eventually do it. And then, as you point out, once the infrastructure is in place, people will scoop off the income and use it to try other even stupider things. Though no doubt the original investors’ll lose their shirts.

  24. More on the light sail. It depends what you regard as a light sail. Photon pressure would be very tiny because photons have almost no mass. So a very large sail. For payloads in the ton range, multiple kilometers across for extremely small amounts of delta V. But one can also use the solar wind. Which is far heavier particles. Protons & alpha particles. Thus a much smaller “sail”. Except it’s just something that intercepts the particles & benefits from the momentum transfer. You’re unlikely to able to tack with it.

  25. It might work but it’s counter intuitive. The solar wind would push your craft away from the sun. The opposite direction to which you want to go. But this is orbital mechanics not earthside. Your craft has velocity of a circular orbit at the asteroid altitude from the sun. Raising the orbit increases gravitational potential but reduces orbital velocity. Since you’ve reduced orbital velocity, although you’re being pushed away from the sun you’re also now falling towards it. So the orbit becomes an ellipse. Raise the aposol of the orbit enough & the perisol intercepts the earth’s orbit. Unfortunately this is the highest velocity point of an elliptical orbit. Very difficult to catch as it goes past. The question is, can you use the solar wind pressure to slow you in the descent phase of the ellipse? Like a parachute. This is where my understanding of orbit mechanics fails. Sounds possible but.

  26. In case you don’t understand why accelerating makes you go slower, this is curved space not flat space. It’s angular momentum you lose. You’re straight line acceleration is exchanged for gravitational potential. In flat space terms, height from the sun.

  27. Another way of understanding it: In flat space one can talk about stationary points & straight line distances between them. In curved space there is only one stationary point in the solar area. The exact centre of the sun. At any other point, unless it has its own velocity/vector, it’s assigned one. It immediately starts falling towards the centre of the sun. So to be “stable”, anywhere else, requires a angular velocity & vector. Their are no straight lines. Travel between two points will always be a curve.

  28. The exact centre of the sun.

    Pendulous pendantry, but everything, including the Sun, swings around the centre of gravity of the solar system (which is mostly the centre of gravity of the Sun and Jupiter). That barycentre spends more time outside the Sun than inside.

Leave a Reply

Your email address will not be published. Required fields are marked *