One night, as I was putting my daughter to bed and waiting for her to fall asleep, I tried to think of some new markets for space utilization.
We often hear about attempts to find industrial uses for microgravity for growing crystals, for purification of electronic materials (which is an actual thing with ACME Advanced Materials: http://www.a2-m.com/ ), maybe growth of certain metal foams, etc. However, in space, you’re in both a hard vacuum and not physically resting on anything, so you can spin up something, and it will simply keep on spinning (stably, if you spin it around the correct axis) nearly indefinitely without any additional energy input and no wear on bearings or anything. So in fact, you can get basically any gravity level you want, including HYPERgravity, nearly for free.
What are the applications of this?
The most obvious one I can think of that has the biggest market potential is isotopic enrichment of Uranium-235 for nuclear fission fuel. The world demand for electricity is about 10^20 Joules electric, and the price of Uranium fuel is about half a cent per kWh. About 40% of that is the cost of separation, with a Separative Work Unit costing around $100. So enrichment cost about 5*10^-10 dollars per Joule of electricity. That gives a world market for up to $50 billion for separation if we used just nuclear. If 10% of our market is nuclear, then $5 billion. Given 10^14J/kg of fissionable fuel and cost of 5*10^-10 $/J, then you have $50,000/kg of Uranium. Knock that down to 20% due to thermal to electrical conversion (power plants are usually better than that), and we’re at $10,000/kg. If you can get launch costs down to $50/kg, then it might be worth doing this (because you’re launching natural isotope ratios of Uranium to make the math easier). But interestingly, near-pure U235 is something that probably WOULD be economically worthwhile to export from Mars or Moon or asteroids (processed in orbit).
One can imagine other uses for isotopic separation, like lithium-6-enriched metal alloys. Lithium 6 is about 15% lighter than natural lithium. The best conductivity-to-mass-ratio wires (other than superconducting or microscopic graphene or nanotubes, etc) at room temperature is Lithium-6, nearly 4 times as good as copper. But that’s a much smaller market.
EDIT: I want to add some more realistic figures for cost, etc. There is currently perceived to be a glut of separation capacity, so the SWU price is still just $82 or so. But there’s also kind of a near-monopoly (quadropoly or something?) among separation providers, and the US is still using crappy gaseous diffusion plants which are super inefficient. So there may be a good argument for doing it anyway as a form of avoiding a sort of cartel arrangement. Also, since natural uranium is so poor in U235, it may actually make sense to pre-enrich the uranium before launch so that you don’t have to launch as much of it. Suppose we’re trying to make 95% U235-enriched Uranium (quite highly enriched) and want our “tails” to contain just 0.1% U235 (vs 0.7% naturally). We want to maximize the number of SWUs we do for a given launched mass. I’ve found that occurs at about 25%-U235 pre-enriched. 5.23 SWUs per kilogram of uranium. According to this calculator: http://www.wise-uranium.org/nfcue.html
(I could have used the full expression for SWUs and massaged it with calculus to give the actual maximum, but I’m getting lazy.)
Multiplying that by the cost per SWU (about $82), we get about $430 per kilogram. In other words, the market value of the work we can do separating that Uranium, if our orbital isotope enrichment goes well,
is about $430/kg. So we probably need launch prices to be down around $100-200/kg for this to really work. But that’s not unreasonable, and there’s also some value in having an independent capability to do this.
….Then again, the elephant in the room here is that we’re talking about launching tons of already-highly-enriched uranium (enough to make a crude fission bomb) and recovering VERY highly enriched uranium. Enriched so much that we’d better be careful about how much we have together at one time. Kind of goes without saying that there’d be political opposition to such a scheme! But still, it would be another space market.
(And I redid the numbers for more power-plant-grade 4.5% low enriched uranium given natural 0.711% feed and 0.1% tail… It’s about $119-worth-of-work/kg launched, so not actually as bad as I thought, and isn’t super highly enriched and so politically is more feasible, since you’re not just shipping bomb-grade material around.)
…and all this is pretty irrelevant if you start breeding your fuel from natural uranium and thorium, which probably makes sense in the long term.
Chris Stelter
Latest posts by Chris Stelter (see all)
- How much mass can we put in orbit before running into atmospheric constraints? - July 19, 2020
- Adding an Earth-sized magnetic field to Mars - June 18, 2020
- A human tribe is a Von Neumann probe - May 24, 2020
Brilliant.
I wonder if it is likely that the enrichment could be done in significantly less steps than current practice. I am guessing that a ground based centrifuge has a maximum speed of well under 1,000 fps, while one of yours could easily triple that using basically tether tech. Also, it would seem that much larger facility sizes would be practical. It seems possible that the hundreds of small heavy units on the ground could be matched by a handful of larger but lighter units in vacuum.
On further thought, Ken had it right. You may have just discovered one of the killer apps.
I’m not sure doing this in space has any significant advantage. My understanding is that the hard part of designing a centrifuge is making it strong enough that it doesn’t tear itself apart. Space doesn’t really help with that. You do save the energy you’d otherwise use on keeping the centrifuges rotating but you have to weigh that against the energy needed to put uranium in orbit.
Andrew,
Obviously I disagree. Consider a carbon composite hourglass shape in free fall. The tip limits are basically tether limits of about 3,000 m/s or so. Very simple in concept at least.
That’s about an order of magnitude higher than you can get at sea level without heroic engineering efforts with sonic limitations in air, or massive vacuum structures. Even if only a factor of three is functionally available, exponential functions make the microgravity units far more effective each as well as simpler in concept.
The following is the result of a Wikipedia binge. I am not an expert on any of this.
I think uranium enrichment is way out of the league of what a newspace startup might be aim for, at least initially. It would be nice to low barrier to entry. It’s a good idea to establish the smallest viable profitable company early on, and grow it into a multi-billion dollar industry once all the fundamentals have been proven on a much smaller scale. No one is going to invest that much startup capital until the tech is proven, so any niches that a company can fill during the R&D phase would help dramatically.
Even if such a company never went big, it seems like having lots of telemetry data on even small spinning crafts might be useful for rotating space tethers or artificial gravity research, both of which would be extremely useful to have some day.
Someone may want to check with some ultracentrifuge[1] experts, or go to some conferences, to see if there is a demand for higher g applications than can be achieved on earth. (Assuming much higher g’s can plausibly be achievable in space. Or is the main advantage that you can set something spinning and leave it for months without power input?)
For example, something like plutonium might have less red tape than uranium, and have the bonus of being extremely useful for RTGs in spacecraft. The Pu-238 Wikipedia article[2] says “Reactor-grade plutonium is not useful for producing Pu-238 for RTGs because difficult isotopic separation would be needed.†I’m not sure, but this might be implying that g-force is the limiting factor. There are only a couple methods available for isotope separation,[3] so that seems at least somewhat likely. Unfortunately Wikipedia doesn’t list a source on this, so I couldn’t easily find more details. Anyone know if this is plausible?
Also, the ultracentrifuge article[1] says this: “There are two kinds of ultracentrifuges, the preparative and the analytical ultracentrifuge. Both classes of instruments find important uses in molecular biology, biochemistry, and polymer science.†Analytical ultracentrifuge applications may be appealing for a space startup, since it doesn’t necessarily require returning the centrifuged material, but only observing it during or after being centrifuged. Perhaps a CubeSat sized centrifuge would be possible, which could be allowed to burn up in the atmosphere after running a customer’s experiment.
Some advanced materials rely on ultracentrifuges for purification. Carbon nanotubes, for example, must be sorted by diameter for many applications.[4] Ultracentrifuges are one of the best available ways to do this, and I suspect that they are the limiting factor in purification. Perhaps this is the sort of thing that absurdly-high g’s for long periods would help with?
[1] https://en.wikipedia.org/wiki/Ultracentrifuge
[2] https://en.wikipedia.org/wiki/Plutonium-238#Production
[3] https://en.wikipedia.org/wiki/Isotope_separation
[4] https://en.wikipedia.org/wiki/Carbon_nanotube#Application-related_issues
The idea isn’t that you’d be able to attain higher gee-loads. Both on Earth and in orbit, you’re limited by material strength. But the difference is it takes zero energy to maintain the speed in orbit. And you don’t need a carefully balanced centrifuge to avoid huge, destructive vibration like you would on Earth. And you also don’t need to maintain energy and wear and tear for the high speed bearings. And you don’t need an expensive vacuum jacket with fancy moving vacuum seals. You also have access to sunlight for power and “free” deeply cryogenic cold (provided you’re properly shielded and in the right orbit).
If in Earth orbit, you can even use the Earth’s magnetic field to spin up (using a little solar energy and a coil) and spin down (generating energy). In interstellar space, you can use the Sun’s weaker (at 1AU) magnetic field, but that’s 1/10,000th as strong (usually).
Centrifuges require 60kWh per SWU (Separative work unit, about 4000 of which are needed per kg of very highly enriched U235), so the main advantage is the other aspects.
I suspect the centrifuges and associated plant would mass a lot more than their annual output of refined isotopes.
Chris,
I think you underestimate the quality of your idea. Working in a natural vacuum with less requirements for fancy bearings and seals allows for higher performance equipment that is still simpler in construction.
I did a small bit of checking on my initial assumptions and found them to be off. Ground based units apparently run up to 700 m/s or so in a vacuum jacket, or about double my assumption. Higher speeds are definitely associated with higher efficiency. Design of a freefall in vacuum unit that can reliably operate at double the speed of ground based units is relatively easy with modern composites. That roughly quadruples the separation effectiveness per stage. A simpler unit four times as effective is a major deal.
I updated with this:
EDIT: I want to add some more realistic figures for cost, etc. There is currently perceived to be a glut of separation capacity, so the SWU price is still just $82 or so. But there’s also kind of a near-monopoly (quadropoly or something?) among separation providers, and the US is still using crappy gaseous diffusion plants which are super inefficient. So there may be a good argument for doing it anyway as a form of avoiding a sort of cartel arrangement. Also, since natural uranium is so poor in U235, it may actually make sense to pre-enrich the uranium before launch so that you don’t have to launch as much of it. Suppose we’re trying to make 95% U235-enriched Uranium (quite highly enriched) and want our “tails” to contain just 0.1% U235 (vs 0.7% naturally). We want to maximize the number of SWUs we do for a given launched mass. I’ve found that occurs at about 25%-U235 pre-enriched. 5.23 SWUs per kilogram of uranium. According to this calculator: http://www.wise-uranium.org/nfcue.html
(I could have used the full expression for SWUs and massaged it with calculus to give the actual maximum, but I’m getting lazy.)
Multiplying that by the cost per SWU (about $82), we get about $430 per kilogram. In other words, the market value of the work we can do separating that Uranium, if our orbital isotope enrichment goes well,
is about $430/kg. So we probably need launch prices to be down around $100-200/kg for this to really work. But that’s not unreasonable, and there’s also some value in having an independent capability to do this.
….Then again, the elephant in the room here is that we’re talking about launching tons of already-highly-enriched uranium (enough to make a crude fission bomb) and recovering VERY highly enriched uranium. Enriched so much that we’d better be careful about how much we have together at one time. Kind of goes without saying that there’d be political opposition to such a scheme! But still, it would be another space market.
(And I redid the numbers for more power-plant-grade 4.5% low enriched uranium given natural 0.711% feed and 0.1% tail… It’s about $119-worth-of-work/kg launched, so not actually as bad as I thought, and isn’t super highly enriched and so politically is more feasible, since you’re not just shipping bomb-grade material around.)
…and all this is pretty irrelevant if you start breeding your fuel from natural uranium and thorium, which probably makes sense in the long term.
Perhaps we are limiting ourselves by saying hypergravity, but perhaps variable gravity? Say you want to make an advanced composite plate/fabric that is a fancy layer cake with infused materials, that is manufactured in combinations of low grav vacuum deposition and high grav direct application/settling. Like advanced macro/micro lattice structures then immersed/embedded in various bulk materials that are differentially applied over the cross section.
Ah, those metal foams using carbide spheres. Use hypergravity so the spheres settle into an optimum dense packed configuration. Military is always willing to pay for fancypants armor for some people.
Turbine blades are differentially solidified and might benefit from being forged under hypergravity conditions.
“Kind of goes without saying that there’d be political opposition to such a scheme!”
Yah, it’s prolly a good thing the Challenger and Columbia weren’t loaded with tons of enriched uranium!
How about, instead of exploiting microgravity, you exploit vacuum?
Do isotope separation magnetically, using ion beams. Use the Earth’s magnetic field so you don’t have to build magnets. You just need an ion beam source and, some distance away, a pair of collector buckets. You might need to vary the beam energy/direction to compensate for changes in the magnetic field as the facility orbits the Earth.
Addendum: I understand the only reason the calutrons at Oak Ridge could produce enough uranium was that they evaded the space charge limit on the beam current by neutralizing the beam with electrons from ambient gas molecules in the imperfect vacuum. This effect, discovered experimentally during the war, was not known in Germany or Japan, where they decided electromagnetic separation could not work.