• Google
    This Blog Web

October 2011

Sun Mon Tue Wed Thu Fri Sat
2 3 4 5 6 7 8
9 10 11 12 13 14 15
16 17 18 19 20 21 22
23 24 25 26 27 28 29
30 31          

RSS Feed

Bookmark and Share

Email Feed

  • Powered by FeedBlitz

« Enlightenment Under Fire | Main | No One is Watching »

August 05, 2006


Feed You can follow this conversation by subscribing to the comment feed for this post.


One thing I never understood about the convergent nanofac designs, is do you have dangling bonds all the way up?

Tom Craver

You mean to attach the parts at larger scales?

I believe the answer is "no - larger parts connect mechanically".

If you want to do objects that are a single molecule all the way up, you'd probably need to build it up directly, instead of building it in cubically sub-divided parts to be assembled. I presume that would be slower, as you'd be limited to only building with the manipulators you can fit within that surface, instead of building parts within a 3D volume and feeding them forward to be quickly attached to the construction face.


I could never see how it could ever be possible to have dangling bonds all the way up in the convergent nanofacs. I have seen mention of a peg concept. To my understanding, at small enough scales it is possible to have pegs that allow you to help guide two dangling bond surfaces to connect and bond. There would of course be many complex reconstructions that would occur the idea being that the pegs would force the final rest state of the new part to be roughly what you wanted. This approach must have some limit, and at larger scales must fail. So then, there are mechanical connections proposed for larger assemblies.

Does the same then hold true for the planar designs? It seems to me if all the blocks being deposited are small enough, then this approach would allow you to build one giant diamond.

I wonder how much practical difference there is to the properties of the finished products made by the two nanofac designs. I also wonder what promoted Drexler to change his mind on which to use.

Chris Phoenix, CRN

There's not only one kind of convergent assembly nanofactory. If I understand right, Drexler's picture when he wrote Nanosystems (1992) was to build odd-shaped pieces and assemble them into larger odd-shaped machines by means of fairly advanced industrial robotics. In this case, much of the larger-stage joining would be placement rather than fastening.

The cube-of-cubes product design (a variant of which is shown in the picture above) was published by Merkle in the mid-90's. This was a simple approach, meant to show that *something* could work. I continued that approach in my Primitive Nanofactory paper. In this approach, it's fastening all the way up.

Planar assembly is based on the insight--in hindsight, very simple and elegant--that rather than passing bigger and bigger cubes forward, you can take the smallest layer of cubes and "pass them forward" directly onto the product. There's no need to go through the intermediate cube stages. This also requires fastening.

In all these cases, things could be fastened either mechanically or chemically. I don't think there's a size limit on chemical joining. A planar surface of any size should be alignable within a fraction of an atomic diameter by spacing alignment pegs over the surface. If it's too mis-aligned for the pegs to go in the holes--for example, due to dissimilar thermal expansion--then a few larger pegs with higher insertion tolerance could be used to pull the surfaces into sufficient alignment for the small precise pegs.

An exception to size-independence is if joining goes better when the surfaces are placed at an angle, joined along an edge, and then allowed to come together slowly. A non-small planar surface will come together flat-on, unless the surface can be warped during assembly.

It's not yet known whether/which dangling bond surfaces will make atomically precise junctions, and which will make a covalent hash that would be, at a guess, equivalent in strength to today's polymers. Since nanoscale mechanical fasteners can preserve more than 50% of the strength of the bulk material, I'm not too worried about whether dangling bond surface joining will work. It'd be nice if it does, but not a showstopper if it doesn't.


The comments to this entry are closed.