[extropy-chat] Space elevator numbers III

[Eugen Leitl]

On Sat, Feb 17, 2007 at 12:30:24PM -0500, Keith Henson wrote:

> If you are at the pole, you can't leave them on the ground.  And how are 

It seems to be just the rim (assuming, it is there at all). I presume
the rim has an inclination. If it doesn't, one has to lift off the sheet
glass product from the regolith. There might bee too much iron and nickel
metal particles in the glass, so one might have to purify the regolith 
magnetically. Apparently other people have gone to greath lenghts
http://www.moonminer.com/Regolith_refining.html already, though this
looks like it's inspired too much by terrestrial approaches, with aqueous
processes, massive complex structures.

If there is no permanently illuminated polar crater rim than I would
just go to the lunar equator, where insolation at noon is vertical, 
but where one has to shut down for two weeks for the night.

> you going to collect the current from the PV surface?

Just like it's being done in commercial amorphous Si cells.
I recall you mentioned the difficulties about gathering current
from very thin silicon surfaces -- it is not a problem with a
fractalish, dendrite type of surface electrode.
 
> If you want to make metals, you need to sort out the oxides before you 

Not really. Though it *is* possible to do a lot of separation of fine-grained
material. Just melting everything and electrolysing the mess will give you
oxygen and a number of alloys and pure metals/elements, most of which are
not miscible, so you separate them at this stage.

> reduce the metals.  There is lot of aluminum in lunar rock, but getting it 
> out as Al2O3 is going to be a major effort.  The Hall process to reduce 
> aluminum not only requires 99% aluminum oxide, but uses huge amounts of 
> carbon, which is burned up at the anodes.

Most terrestrial processes are not applicable to the lunar environment,
and vice versa. For instance, nobody would consider fractional destillation
with solar ovens on Earth. On the Moon, it's a potentially very useful
and workable process. 
 
> http://en.wikipedia.org/wiki/Hall-H%C3%A9roult_process
> 
> It takes sorting very large amounts of regolith with a magnet, but there is 
> a bit of reduced iron from iron containing meteorites.

Yes, getting the iron and ilmenite with a magnet out would be good.
 
> Off hand I don't think there is much in regolith that hydrogen is going to 
> reduce.  Please list for my edification.

Iron, titanium, oxygen.
http://www.nas.nasa.gov/About/Education/SpaceSettlement/Contest/Results/2002/Aether/Construction.htm
http://fti.neep.wisc.edu/neep533/FALL2001/lecture10.pdf

Apparently there's a lot of implanted hydrogen and some sulfur
already, so one has to bake the stuff out to get rid of native
reductive equivalents.

I don't know the details, since this is not my field. But
clearly this is all about details, and improvisation, to find
the right process and the right gadget for the location.
There is considerable variation in the terrain and mineral
composition, so some areas are special (high-titanium, for instance,
or contain massive amounts of implanted Ni-Fe from impactors).
 
> >(closed-circuit water electrolysis)
> >of regolith, and magnetic sorting. Electrolysis in the melt.
> >Fractional destillation. Preparative mass spectroscopy.
> 
> http://en.wikipedia.org/wiki/Calutron

I know about that, of course. But what I meant was something much simpler, 
like http://en.wikipedia.org/wiki/Quadrupole_mass_analyzer

A tennis court or a square mile of these might produce small amounts
of rare, but useful elements. They're tunable, of course. I don't know
whether this would work, but there are some smarter choices in design
space, and some that will make it hard for you. We need to map out
the problem space in order to find out which is which. A lot of it
is experimental. Some of it can be done only in-location. It's a long-term
project, and technology progresses quite rapidly today -- I would
think automation would be the cost-cutter, and allow you to produce
semiautonomous seeds which can bootstrap at lightminute latencies for
control (more guidance, in this case).
 
> It not blind though.  Chemistry is a very well understood subject.

Yes, but e.g. there is not much chemistry in the electrosteel process.
There's plenty of design decisions, though, which are not obvious and
many are not predictable (numerical simulation can be useful, of course), 
until one tries to build a system. For instance, titanium chemistry is
very well known, but http://www.nature.com/nature/journal/v407/n6802/full/407361a0.html
is reasonably recent. You certainly can't do the Kroll process on the Moon,
so something like this appears a straightforward alternative. 
 
> >Because this is expensive, you have to scale down size, and do most
> >of the prototype work Earth-side, in lunar simulators. NASA has just ordered
> >a large batch of simulated regolith, and UHV chambers where you could
> >walk in are expensive, but not nearly as expensive as actually soft-landing
> >a kiloton of hardware on Moon surface.
> 
> Ah, we are already up to a kiloton.

No. I was pointing out that you might not have to soft-land a kiloton
if you do the hard work before Earth-side. How much total mass the
seed (replication closure slightly over unity) will have depends
very much on the technology. With nanotechnology, some few ten cubic
microns, or at least few mg-g might be enough. There's a continuum
up from there through mesoscale to megascale. Anything involving
launching humans is distinctly macroscale to megascale. Here a kiloton
is just a warm up. Nobody has the money anymore.

Basically, in absence of large-scale projects we will have to make
do with small-scale, smart projects. There's really no alternative
to teleoperation, automation, and self-rep, which is a continuum 
(closure 0...>>1). 

> They are closely related.  See the paper Eric Drexler and I wrote on vapor 
> phase fabrication.  In that case the apparatus was able to deposit its own 
> mass in metal every 8 hours.

That's one heck of a productivity. Initial bootstrap can be slow, e.g.
transfer time Earth-Moon with an ion drive takes about 6 months. It can
take a decade or two to achieve a closure of over unity. But since this
is a positive autofeed process, things can and will explosive later.
 
> To this day I have a chunk of sheet metal I made in a high vacuum chamber 
> by vaporizing aluminum with a 10 kw electron beam.  I might add that 
> getting rid of waste heat is often more of a problem than energy.

I'm really impressed. You certainly know about this stuff far more than
I do. I wonder where you see the difficulties so high you would propose
a tether, which is a far more demanding design space than a lunar
bootstrap, objectively so.
 
> >How many launches into high Earth orbit would it take to get to get your
> >material (counterweight and the carbon nanotube cloth belt) up?
> 
> The counterweight is salvaged space junk.  Brad Edwards thinks you can 
> start with an 18 ton seed cable.  The cable is not cloth and not a 
> belt.  There are no climbers, just elevator cars going up a moving cable.

Do you have a specific online document I we can read? Apparently your approach
is quite different from the usual Rapunzel (let down your carbon braids)
idea. 
 
> Please go back and text search.  Have I mentioned an asteroid as the 
> counterweight?

I've probably missed the URL you posted. Can you repost the link?
 
> I am writing a novel where 99% of the population has uploaded so in the 
> long run I too see a computronium future.  This project is a stop gap 
> measure to bridge between now and full scale nanotechnology.  You can't 
> quit farming now because people will be living on electricity some time in 
> the future.

There are considerable disruptions in our future short-term. There's
already unrest brewing in Mexico over rising corn prices because the
gringos decided to move on to bioethanol destilled from corn syrup
melasse.
 
> If discussing space elevators and power sats is too low tech and too near 
> term to hold on this group I will move it.

No, it is interesting. Please continue here. I'm particularly interesting
on how to improve on phased-array radiator targetting the terrestrial rectenna
approach. There must be publications about this problem I'm unfamiliar with.
They probably predate the world wide web, so online sources will be scarce.

-- 
Eugen* Leitl <a href="http://leitl.org">leitl</a> http://leitl.org
______________________________________________________________
ICBM: 48.07100, 11.36820            http://www.ativel.com
8B29F6BE: 099D 78BA 2FD3 B014 B08A  7779 75B0 2443 8B29 F6BE
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 191 bytes
Desc: Digital signature
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20070217/853e57c6/attachment.bin>