6 ms·

Quoting further, "the mass and scale of the sunshades will also be similar to current terrestrial civil engineering projects such as the Chinese Three Gorges Da

by fernly 11y ago

Quoting further, "the mass and scale of the sunshades will also be similar to current terrestrial civil engineering projects such as the Chinese Three Gorges Dam [20], and require a mass production of coated thin material equivalent to the current world decadal production of aluminium foil." They continue with "Nevertheless, scholarly work has yet to identify a scientific showstopper for its implementation". To me, the idea of designing a practical support framework for a disc of aluminium foil having a diameter of one thousand four hundred kilometers -- a framework that has to be shipped from earth one rocket-load at a time and assembled in space -- would be at least a show slower-downer.

mrec 11y ago
There's a substantial amount of aluminium on the Moon, as well as in some asteroids. Extracting and turning it into sunshade won't be easy, obviously, but when budgets in the tens or hundreds of trillions are being talked about, the options widen a bit.
- JoeAltmaier 11y ago
  Wow you're right! Its apparently a huge thing on the moon: "Aluminum composes 10% of the atoms and 13% of the mass of lunar highland regolith, being the third most abundant element. In the mare basins, aluminum makes up only 4.5% of the atoms and 5% of the weight, strongly suggesting the use of highland feedstocks for aluminum extraction." Pretty much a little solar energy and a crucible and you'd turn out aluminum like mad. Absolutely the right choice for a lunar civilization, instead of lifting structures from Earth's gravity well.
  - fernly 11y ago
    Right! So all we have to do is establish a base on the moon, build a mining and smelting facility, build a fabrication plant, build a rail-gun launcher, operate all three long enough to fling at least 10e7 tonnes [1] of finished aluminum parts toward L1, where we catch them and assemble them. The rail-gun is not likely to be the pinch-point in the process, but suppose that the mining, smelting, fabrication, packaging and transport process can deliver a shipping unit to the rail head every minute, around the clock [2]. That's 10e7 minutes, pretty close to 20 years of continuous operation. This stream of 1-tonne bundles of finished rods and other parts has to cross space to L1. I have no idea what the orbital feasibility of Luna-to-L1 is; I'm just assuming that a rail-gun can impart enough energy to get it there. But, one, if a bundle requires any guidance or course corrections en route, that would mean every bundle would have to be equipped with at least a computer, gyros and some type of propulsion units; and two if it approaches L1 with any significant relative velocity, it needs all that and a serious rocket engine to slow down [3]. So we are not merely flinging passive bales of aluminum parts from the rail-gun, we are launching little one-tonne spacecraft. One per minute. For 20 years. [4] So supposing all that is in place, then you need to have sufficient personnel (plus a swarm of quite intelligent robots) at L1 to receive them, unbundle them, and distribute the parts across a disc that even in the earliest stages is hundreds of fucking kilometers in diameter, and assemble them. For 20 years. There is nothing in this scenario that is inherently impossible. On the other hand, given it currently takes us a decade to design and transport a 1/2-tonne robot to Mars, how many decades would it take to design and build the proposed Lunar infrastructure? On time alone, never mind the costs, wouldn't it be simpler, quicker and cheaper to just stop burning petroleum? [1] the article says 10e7 - 10e8, so we're giving them the low side of a power of 10 here. [2] something like a factory that builds a new economy car every 2 minutes. On the moon, in a vacuum. [3] or it has an ion engine like the Dawn craft and does continuous thrust, yadda yadda, same basic problem. [4] you might say we could recycle the computer and ion engine from each bale, but then you need to figure out how to fire about 10e6 bundles of used engines from L1 back to the moon and soft-land them there next to the factory.
    - chrispeel 11y ago
      Fernly, great enumeration of some of the challenges! I'm completely with you on the difficulty, and agree that this won't happen soon. Even so, 200 years ago the steam engine was just being developed; since then we've had the rest of the scientific, industrial, and information revolutions. So the question is what will happen 200 or 2000 years from now? A sunshade seems plausible on that time-frame, and the paper is one way to begin the conversation about such a project.
    - JoeAltmaier 11y ago
      Here's some help: regolith is broken rock and dust on the ground. So no mining necessary - just earth movers (luna movers?) and a conveyor belt. Probably the first thing you'd build would be - more mining equipment. Let the thing bootstrap itself, including the fabrication plant, the rail-gun and more mining equipment. In 20 years you'd be ready to begin doing the rest of the project, but it would be much easier with the orders-of-magnitude larger infrastructure you'd bootstrapped.
      - sbierwagen 11y ago
        Incidentally, NASA commissioned a study back in 1980 about self-replicating factories harvesting lunar regolith. It's quite good, and goes into excellent detail: http://www.nss.org/settlement/moon/library/1982-SelfReplicatingLunarFactory.pdf http://www.nss.org/settlement/moon/library/1982-SelfReplicat... (Edited by Robert A. Freitas, who went on to write Kinematic Self-Replicating Machines)
  - sbierwagen 11y ago
    Aluminum is similarly abundant in the Earth's crust. Most common element after oxygen and silicon, twice as abundant as iron. Correctly tempered and hardened aluminum is lighter and much stronger than iron. The sole reason aluminum isn't the "default metal" in the way iron is, is that it's so much harder and more expensive to smelt: Electric power represents about 20% to 40% of the cost of producing aluminium, depending on the location of the smelter. Aluminium production consumes roughly 5% of electricity generated in the U.S.[26] I agree that pretty much everything manufactured on the Moon is going to be made of aluminum, for precisely those reasons, but needing a "little" solar power is underselling it a little.
    - JoeAltmaier 11y ago
      But the two things you have on the moon - unrestricted view of the sun, aluminum dust lying around in piles - means its certainly the easiest of the alternatives. That's the point.
      - sbierwagen 11y ago
        It's aluminum-containing dust, not bulk metallic aluminum. All of the aluminum-containing minerals on the Moon have the aluminum atoms bound to oxygen, silicon, etc. Smelting aluminum is expensive because oxygen bonds to aluminum very tightly, and requires a lot of electricity to pull it off.

by fernly 11y ago

mrec 11y ago

There's a substantial amount of aluminium on the Moon, as well as in some asteroids. Extracting and turning it into sunshade won't be easy, obviously, but when budgets in the tens or hundreds of trillions are being talked about, the options widen a bit.

JoeAltmaier 11y ago
Wow you're right! Its apparently a huge thing on the moon: "Aluminum composes 10% of the atoms and 13% of the mass of lunar highland regolith, being the third most abundant element. In the mare basins, aluminum makes up only 4.5% of the atoms and 5% of the weight, strongly suggesting the use of highland feedstocks for aluminum extraction." Pretty much a little solar energy and a crucible and you'd turn out aluminum like mad. Absolutely the right choice for a lunar civilization, instead of lifting structures from Earth's gravity well.
- fernly 11y ago
  Right! So all we have to do is establish a base on the moon, build a mining and smelting facility, build a fabrication plant, build a rail-gun launcher, operate all three long enough to fling at least 10e7 tonnes [1] of finished aluminum parts toward L1, where we catch them and assemble them. The rail-gun is not likely to be the pinch-point in the process, but suppose that the mining, smelting, fabrication, packaging and transport process can deliver a shipping unit to the rail head every minute, around the clock [2]. That's 10e7 minutes, pretty close to 20 years of continuous operation. This stream of 1-tonne bundles of finished rods and other parts has to cross space to L1. I have no idea what the orbital feasibility of Luna-to-L1 is; I'm just assuming that a rail-gun can impart enough energy to get it there. But, one, if a bundle requires any guidance or course corrections en route, that would mean every bundle would have to be equipped with at least a computer, gyros and some type of propulsion units; and two if it approaches L1 with any significant relative velocity, it needs all that and a serious rocket engine to slow down [3]. So we are not merely flinging passive bales of aluminum parts from the rail-gun, we are launching little one-tonne spacecraft. One per minute. For 20 years. [4] So supposing all that is in place, then you need to have sufficient personnel (plus a swarm of quite intelligent robots) at L1 to receive them, unbundle them, and distribute the parts across a disc that even in the earliest stages is hundreds of fucking kilometers in diameter, and assemble them. For 20 years. There is nothing in this scenario that is inherently impossible. On the other hand, given it currently takes us a decade to design and transport a 1/2-tonne robot to Mars, how many decades would it take to design and build the proposed Lunar infrastructure? On time alone, never mind the costs, wouldn't it be simpler, quicker and cheaper to just stop burning petroleum? [1] the article says 10e7 - 10e8, so we're giving them the low side of a power of 10 here. [2] something like a factory that builds a new economy car every 2 minutes. On the moon, in a vacuum. [3] or it has an ion engine like the Dawn craft and does continuous thrust, yadda yadda, same basic problem. [4] you might say we could recycle the computer and ion engine from each bale, but then you need to figure out how to fire about 10e6 bundles of used engines from L1 back to the moon and soft-land them there next to the factory.
  - chrispeel 11y ago
    Fernly, great enumeration of some of the challenges! I'm completely with you on the difficulty, and agree that this won't happen soon. Even so, 200 years ago the steam engine was just being developed; since then we've had the rest of the scientific, industrial, and information revolutions. So the question is what will happen 200 or 2000 years from now? A sunshade seems plausible on that time-frame, and the paper is one way to begin the conversation about such a project.
  - JoeAltmaier 11y ago
    Here's some help: regolith is broken rock and dust on the ground. So no mining necessary - just earth movers (luna movers?) and a conveyor belt. Probably the first thing you'd build would be - more mining equipment. Let the thing bootstrap itself, including the fabrication plant, the rail-gun and more mining equipment. In 20 years you'd be ready to begin doing the rest of the project, but it would be much easier with the orders-of-magnitude larger infrastructure you'd bootstrapped.
    - sbierwagen 11y ago
      Incidentally, NASA commissioned a study back in 1980 about self-replicating factories harvesting lunar regolith. It's quite good, and goes into excellent detail: http://www.nss.org/settlement/moon/library/1982-SelfReplicatingLunarFactory.pdf http://www.nss.org/settlement/moon/library/1982-SelfReplicat... (Edited by Robert A. Freitas, who went on to write Kinematic Self-Replicating Machines)
- sbierwagen 11y ago
  Aluminum is similarly abundant in the Earth's crust. Most common element after oxygen and silicon, twice as abundant as iron. Correctly tempered and hardened aluminum is lighter and much stronger than iron. The sole reason aluminum isn't the "default metal" in the way iron is, is that it's so much harder and more expensive to smelt: Electric power represents about 20% to 40% of the cost of producing aluminium, depending on the location of the smelter. Aluminium production consumes roughly 5% of electricity generated in the U.S.[26] I agree that pretty much everything manufactured on the Moon is going to be made of aluminum, for precisely those reasons, but needing a "little" solar power is underselling it a little.
  - JoeAltmaier 11y ago
    But the two things you have on the moon - unrestricted view of the sun, aluminum dust lying around in piles - means its certainly the easiest of the alternatives. That's the point.
    - sbierwagen 11y ago
      It's aluminum-containing dust, not bulk metallic aluminum. All of the aluminum-containing minerals on the Moon have the aluminum atoms bound to oxygen, silicon, etc. Smelting aluminum is expensive because oxygen bonds to aluminum very tightly, and requires a lot of electricity to pull it off.