[extropy-chat] Re: Self replication again (was MARS: ...

[Chris Phoenix]

Stephen J. Van Sickle wrote:
 > Hmmm....then this brings up the question of how one safely and cheaply
 > disposes of used/defective/outdated devices. Ideally you would want to
 > close the material loop and turn it into feedstock, but as you say it is
 > not obvious that this is simply an inverse operation of construction.
 > Worst case, I suppose, would be some sort of high pressure and
 > temperature decomposition, but the costs would be significant,
 > especially in comparison to the cost of manufacture.  However, it seems
 > a necessary area to investigate when trying to evaluate the cost or
 > effectiveness of large scale MNT.

It's basically a chemical processing operation.  They tend to be pretty 
cheap.  I don't know whether it would be cheap or expensive by 
comparison with manufacturing.  But it would almost certainly be cheap 
relative to the value of the product.

Mike Lorrey wrote:
 > That is a third problem entirely, and likely the simplest one to deal
 > with. If you know the design of the device you are disposing of, you
 > just disassemble it into its constituent parts, toss out the defective
 > ones, and put the rest into parts bins.

There are several practical problems with this.  Radiation damage may 
change the molecular arrangement.  Contaminant molecules on the surface 
may be hard to deal with.

 > With the first problem you are dealing with a 'ore' of rather
 > high-entropy material in terms of isotope dispersion. Picking each atom
 > out one at a time, shaking it, and sorting it into the right bin is a
 > time consuming but rather simple process.

It's only simple if you already have solved the problem of building 
nanomachines.

 > The second problem is the toughest, building something that is highly
 > structured atom by atom from a design.

I don't see what's so tough about this, at least for simple chemistries 
like diamond.  Diamond should be buildable with just a few operations, 
parameterized--and the parameterization is part of the design, and 
should be simple to derive from a simple blueprint.

Diamondoid as implied in Nanosystems--carbon backbone, but with lots of 
other elements substituted or inserted in idiosyncratic places--may be 
tougher.  But it appears likely that we can build a molecular 
manufacturing capability with just diamond and maybe graphene/fullerene.

Chris

-- 
Chris Phoenix                                  cphoenix at CRNano.org
Director of Research
Center for Responsible Nanotechnology          http://CRNano.org