On 5/21/2011 6:58 PM, Chuck Bowling wrote:
On Sat, May 21, 2011 at 1:37 PM, Ash <[email protected]
<mailto:[email protected]>> wrote:
On 5/21/2011 10:42 AM, Chuck Bowling wrote:
Interesting article. I scanned more than read but it seems like a
valid architecture for reconfigurable molecular systems.
I just wonder if it might be overkill. It seems to me like a more
versatile system would be a relatively simple 3D printer. We
already have 3D printers that can print any polymer 3D part. I
would think that the technology should be transferable to
nanotech. We could literally print anything atom by atom.
I'm afraid that the principle of printing the way we do today
would be terribly inefficient/ineffective on nanoscale 3d structures.
Why would a few billion nanites all working together to build a 3d
structure be inefficient?
For your average product we could nanoassemble polymers for various
molecules, create a mix and maybe use UV or another imaging method if
that's what you meant. 3d printing could be vastly superior with
nanoengineered materials and controls, probably even superior to today's
industrial products, but 3d printing is dog ass slow (another
improvement area. I'm imagining a 1billion dpi mems diamond grid
rastering out surfaces, maybe it would look more like a bunch of tiny
eyeballs (concentrating lenses) scanning a surface with growths
following it's path and a stream of bots/parts flowing up the surface
along printed distribution lanes/veins.
I don't mean that it necessarily would be inefficient, and there is
something to be said for building things in layers, much of those
concepts should be sufficient for assembling larger objects but I think
it would work more like Lego blocks where a billion nanites build chunks
and have some at the inter-faces to bond the chunks and do alignment.
Then larger assembly bots put the pieces together with QC beacons
programmed into the interfaces for guaranteed alignment, conditioned
environments for the work to take place at varying temperature and
chemicals in the atmosphere, atmospheric pressures needed to ensure
quality. I can't think of a way to get from atomic assembly to finished
macro-scale object (computer chip or something) without many levels of
abstraction each level probably decreasing in overall quantity and
diversity of bots for a project. There would be a need for correction of
devastating forces at those scales like environmental vibrations and
other dynamic forces. That is, in cases where nanoengineering shines.
Man I love this stuff.
Consider machining or layered deposition techniques, there are
angles that need to be locked in by something maneuverable on all
axes at a very small scale, probably at times the assembler will
need to trap itself within the structures in order to complete
them well.
We already have nanobots that move in 2d over an atomic lattice. It
should be a fairly simple process to add a 3rd dimension for layering.
As to the assembler trapping itself, I'm not a molecular engineer but
I don't see the problem in transferring techniques for building large
scale parts to the realm of nano-structures. For instance, creating a
hollow sphere doesn't require that the sphere be formed as a single
unit. It can easily be created as two hemispheres and welded together.
Piecing larger parts together is a great way to speed things up. I
hadn't heard about that progress on lattices, very neat. It has been a
very long time since I've actually read anything on nanotech, I should
when I have more time. BTW I need to reread the article I linked but in
my half-asleep state I got the impression that it was more of a
preliminary tech review than a mature solution, there were many gaps.
