Spike Jones wrote:
> ok now go back down the scale. we are accustomed to getting power
> from violent chemical reactions such as octane and air, but a plant can
> get all the energy it needs from photosynthesis. plants on our scale
> generally need a lot of leaves, but single celled plants have so little
> chlorophyll they sometimes dont even look green, and yet they seem
> wildly active under a microscope. chlorophyll creates sugars, which
> break down in a much gentler reaction than oxygen combustion.
> so, if the lesson of nature is not misleading, we dont need nanonukes,
> or even nanoscale combustion. the square cube law would suggest
> that nanobots should be able to create all the energy they need using
> chlorophyll and the fusion reactor nature has already generously
> provided: the sun. right?
As far as making small robots walk, fly, move things around, and so on I think you're right. Solar power should be fine for anything that has access to it, and you should be able to fuel thin layers of nanobots that way with a moderate amount of specialization (solar collector bots on top, which produce fuel molecules consumed by the bots further down).
Once you start talking about nanotech fabrication that changes things. I think Drexler did some calculations on this based on the energy required to break atomic bonds, and the energy requirements turned out to be pretty high. Tearing random matter apart into its constituent atoms is not a very efficient way to make things - which is one reason why nanotech factory systems are likely to build things faster than free-roaming replicators.
Solar-powered nanobots should still be feasible, but they would probably operate on the same time scale as plants (better efficiency and higher energy requirements should come within an order of magnitude of balancing out). That's fast enough to be quite useful, but not fast enough for some of the applications that have been considered - especially for military use, or for really large-scale construction.
Billy Brown, MCSE+I