• 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

« Dry Nano, Wet Nano | Main | Give Me a Nanohand »

March 15, 2004


TrackBack URL for this entry:

Listed below are links to weblogs that reference Molecular Gearheads:


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

Janessa Ravenwood

Hmmm...think I'll save a copy of that article.

Richard Jones

These molecular motors are spectacular and inspiring. However, I'm not sure whether the view of them as mechanical is helpful except at rather a superficial level. If you go into the maths of how they work things look very alien. I'd classify something as mechanical if its components obey Newtons laws - F=ma etc. But the governing equations for the components of molecular motors are Langevin equations; rather than
F = ma
we have equations looking like
F=fv + eta(t)
where f is a friction coefficient, v velocity and eta(t) is a random, fluctuating force with a Gaussian distribution.

The key point is that because the operating physics in the wet, nanoscale world is very different to what we are used to in the macro world our mechanical intuitions aren't always helpful.

Chris Phoenix, CRN

Give us credit for at least a little physical sophistication! By "mechanical," we're not saying protein motors will follow F=ma. We're saying that they generate and transmit force--they can drag stuff around.

More than that, for at least some of the motors, it looks like they take exactly one step per ATP, so their final position is very predictable regardless of the gyrations they go through to get there.

When it comes time to develop systems based on these (or other) motors, we'll certainly have to learn engineering beyond F=ma. But will it be any weirder than fluid dynamics? Even supersonic fluid dynamics? Will it be so weird that we can't learn to engineer in that domain? No, it won't be that weird.

And remember that even F=ma wasn't always intuitive, or necessary: it wasn't invented until many centuries after the Romans, who were certainly great engineers in some domains. Conversely, almost a century after Tesla, significant advances are still being made in electric motors.

Molecular manufacturing shouldn't require pushing any performance envelopes. We won't have to understand the nanoscale fully in order to engineer working mechanochemical systems and a few basic products.



I believe the reason for the eta(t) term is because molecular systems are constantly vibrating because of heat energy.

The work of breaking and making of covalent bonds using STMs and other techniques have all been done at cryogenic temperature and UHV environment, precisely to avoid this thermal "noise". I don't think this approach is going to get us very far.

It does appear that biochemical reactions (ribosomes, ATP synthesis) does occur based on "mechanical" priciples. Molecular mechanical systems based on priciples incorporating the thermal noise in ambient conditions do appear to be possible. Such a nanotechnology could most quickly be developed by "reverse engineering" biology, combined with pure scientific investigation of molecular mechanics in ambient environments.

However, such a nanotechnology strikes me as being biomemetic rather than "diamondoid" in nature. Which, in turn, suggests that the bio-memetic approach is still the shortest, most effective approach to developing a comprehensive "nanotechnology".

The other thing about biology is the hierarchy of biological structure and functionality. There are molecular machines (such as ribosomes) that make up larger scale systems (such as cells) which make up even larger scale systems (multi-cellular organisms). There are several layers of hierarchy with lots of redundancy built in, and the whole thing still uses a certain amount of diffusion-driven chemistry. All of this based on solution-phase chemistry. Solution-phase chemistry offering a far greater range of molecular machines and reaction-mechanisms than vacuum-based (hypothetical) processes.

My problem with Drexler's concept is not so much that I object to the notion of mechnically driven processes on the molecular level. Its the notion that a hypothetical vacuum-based "machine-based" chemistry. I still don't think this is possible and even if it is, its not clear to me that it will have the functionality and the capability to make the broad range of materials that solution-phase approach can.

I still think that, at least on Earth, solution-phase "nanotechnology" is the best, most realistic approach.

Chris Phoenix, CRN

Kurt: Yes, heat energy causes constant vibration. Despite this, it appears that nanoscale robotic systems with an overall stiffness of 10 N/m between chemical tip and workpiece can be built out of diamond (once we can build diamond). And it appears they could do reliable chemistry even at room temperature. The calculations are all in Nanosystems.

Hierarchy is just as available to engineering as to biology. Software engineers know all about hierarchical design and function.

The flexibility of vacuum-based machine-phase chemistry has not been proved. It would certainly make a different class of chemicals than solution-phase chemistry. But the class of accessible chemicals probably includes 3D covalent lattices like diamondoid and alumina. With very little effort, quite a few diamond-forming and hydrogen-transferring reactions have been tested in simulation. The chance that nothing like them will work in practice seems low.

Even if it turns out that diamond is chemically inaccessible, and vacuum chemistry isn't workable, there's still space to improve on biology in the solution phase. Biochemistry generally needs to be efficient and reversible. Engineered nanomachines don't have to be so efficient, and can use irreversible chemistry to make stronger materials.

But when we're planning for the possible consequences of molecular manufacturing, it only makes sense to consider diamond nanomachines a possibility. Vacuum chemistry is different from biochemistry, but it's no weirder than e.g. flame chemistry. If you think vacuum chemistry isn't possible, it's not just your opinion against Drexler's. It's your opinion against Drexler, Merkle, Freitas, and tens of thousands of computer-hours of simulation. At this point, we can't afford to bet against them--the stakes are too high and the odds are nowhere near good enough.


The comments to this entry are closed.