• 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

« Measuring Sustainability | Main | Before the Awakening »

September 10, 2005


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


Chris, I read your paper about building a nanofactory. It's really impressive. Before I've read it I thought that MNT was more then a century away but now I think different.

What do you think about getting the foundations going? With AFMs and sub-angstroem REMs we should already have the basic tools to build and debug first simple nanomachines like those Drexler describes. Are there any experients in this direction you're aware of?

A single AFM is probably to slow to create a machine out of some billions atoms. But we could use some thousands of them in parallel to create a simple prototype of a first working assembler. If one AFM would place only 2 only atoms per minute, 1000 could place 1 billion atoms in a year. Those thousends of parts could then assembled with AFMs also.

The first assembler would probably be smaller then the one you need to create the nanoblocks you described. But even if it only can create structures in the 10nm range with place-rates of some 10000 atoms/s, it should be no real problem to use this assember to duplicate itself and then create bigger assemblers by them later.

So for an investment of around a billion $ we could start MNT few years. But obviously that hasn't happen yet, whats do you think is the reason?

Chris Phoenix, CRN

Carl, I suspect that with automation, an AFM could place several atoms per second. Maybe more. And AFM bootstrapping as you describe is just one approach. Freitas and Merkle like it. Drexler likes an approach that starts with quasi-biological molecules. Hall likes increasingly small micromachining. I change my mind about once a year as to which approach will be most efficient.

Until a few years ago, a lot of people denied that MM could work at all. Professors even told their students not to read Nanosystems. Seems strange today, but that's how it was.

There are still people who say it's too early to try, or we should approach it incrementally, or it hasn't been proved that it's worth doing.

So it's not a surprise that MM has been a bit slow to take off. Of course, the foundational work was published back in 1992, and got some attention from at least a few people in the US military and research communities, so it's plausible that there's a covert program. Some observers say they can't imagine there not being such a program already. Others say they've seen none of the signs they'd expect--no supporting non-classified research infrastructure.



I think that the AFM-way has the big advantage that you don't need an additional paradigm-change: The method you described in your paper uses small double-tripod-fabricators which place atoms the same way as an AFM would do. So working with an AFM would give all neccessary informations you need to program a nano-fabricators if you have one. And if you OTOH can't place single atoms with an ATM in the required way then you will probably also fail to place them with a double-tripod-fabricator.

Since AFMs are relativly affordable for some time now, it would be interesting what kind of progress was made. I remember things like writing little letters from single atoms on a surface, but if would be more usefull to create for example a 3d-diamond-cube out of 5^3 C-atoms.

I've seen a simulation to place a single C-Atom in "Nano-Hive" so it seems possible. Are you aware if someone tried that out in reality?

And if it's really possible to place maybe 10 atoms/s with an AFM, then you need only tens and not thousends of them to create a small fabricator in a year (plus a unknown time for prior experimentation and subsequent bootstapping the real nanofactory). That would be relativly cheap and in the range of many companies.


Chris Phoenix, CRN

Carl, an AFM is more limited than a double-tripod fabricator. The AFM has only 3 DOF, a less clean working environment, and serious trouble recharging the tip or even getting a well-structured tip in the first place. Freitas has a proposal that may solve some of these problems and may be try-able today.

So I agree with you that *if* you can do it with an AFM, you're mostly home free. But the problems with the AFM stage of the project mean that other approaches are still in the running.

At some point in the next few years, or maybe today, MM development will in fact be in the range of many companies.


Mike Deering

In the wake of the government's response to Katrina I think we can rest assured that we are doomed.

Dispite warnings and studies and exercises indicating Katrina would be a different kind of challenge than previous hurricanes, the government's innitial response was identical to it's response to Dennis.

A change in government behavior always lags significantly behind a change in the situation it is dealing with. This lag will be disasterous in the case of MM. Our government is not capable of getting us through the transition safely.

To fix this will require changing the very culture of the government. I don't see how we can get there from here. We're doomed.

Tom Craver

Note what *did* work with Katrina:

There was adequate warning. Our weather monitoring system worked very well, mainly because weather is a continuous threat that frequently springs surprises, so they don't get a chance to become lax.

Unfortunately, that probably won't apply to early MNT crises - we can't expect government to anticipate and give timely warn of them, any more than it did 911.

Most of the population got out safely on their own on fairly short notice - personal and commercial transport worked fine for them. Distributed responses can achieve massive results far beyond any direct government action.

Even if the first MNT crisis - say a massive sneak attack - hits before MNT becomes available to people, I'd expect the most effective responses to come from people taking care of themselves. If we can anticipate likely early forms of attack, and widely distribute information on safety measures, it might help.

I'd guess any early MNT attacks will use massive numbers of fairly simple devices making fairly conventional attacks (fire bombing, knocking out electric power systems, polluting water supplies, etc). But those are things people can prepare for fairly inexpensively - smoke detectors and fire extinguishers, small generators, stored water and water purification equipment, etc.

The comments to this entry are closed.