Corwin

This was inspired by an earlier post about future developments. One thing I noticed is that one theme that keeps coming up is that 'nanotech won't do such-and-such.' However, in general people just make that statement without explaining why.

So... will nanotech live up to its basic claims, and if not, why not?

alek0

Depends whose "basic claims". Very few scientists agree with Drexler. Maybe this will explain why:

Scientific American; New York; Sep 2001; Richard E Smalley;

Volume: 285
Issue: 3
Start Page: 76-77
ISSN: 00368733
Subject Terms: Nanotechnology
Robots

Abstract:
Smalley discusses the prospects of building nanometer-scale robots, as well as the worries that such robots may become self-replicating and parasitic, causing great damage to all biological life. Smalley feels that such developments are virtually impossible, due to problems inherent in the creation and movement of nanobots.

Full Text:
Copyright Scientific American, Incorporated Sep 2001
[Headnote]
How soon will we see the nanometer-scale robots envisaged by K. Eric Drexler and other molecular nanotechologists? The simple answer is never


WHEN A BOY AND A GIRL fall in love, it is often said that the chemistry between them is good. This common use of the word "chemistry" in human relations comes close to the subtlety of what actually happens in the more mundane coupling of molecules. In a chemical reaction between two "consenting" molecules, bonds form between some of the atoms in what is usually a complex dance involving motion in multiple dimensions. Not just any two molecules will react. They have to be right for each other. And if the chemistry is really, really good, the molecules that do react will all produce the exact product desired.

Near the center of the typical chemical reaction, the particular atoms that are going to form the new bonds are not the only ones that jiggle around: so do all the atoms they are connected to and the ones connected to these in turn. All these atoms must move in a precise way to ensure that the result of the reaction is the one intended. In an ordinary chemical reaction five to 15 atoms near the reaction site engage in an intricate three-dimensional waltz that is carried out in a cramped region of space measuring no more than a nanometer on each side.

In recent years, it has become popular to imagine tiny robots (sometimes called assemblers) that can manipulate and build things atom by atom. Imagine a single assembler: working furiously, this hypothetical nanorobot would make many new bonds as it went about its assigned task, placing perhaps up to a billion new atoms in the desired structure every second. But as fast as it is, that rate would be virtually useless in running a nanofactory: generating even a tiny amount of a product would take a solitary nanobot millions of years. (Making a mole of something-say, 30 grams, or about one ounce-would require at least 6 x 10^sup 23^ bonds, one for each atom. At the frenzied rate of 10^sup 9^ per second it would take this nanobot 6 x 10^sup 14^ seconds-that is, 10^sup 13^ minutes, which is 6.9 x 10^sup 9^ days, or 19 million years.) Although such a nanobot assembler would be very interesting scientifically, it wouldn't be able to make much on its own in the macroscopic "real" world.

Yet imagine if this one nanobot were so versatile that it could build anything, as long as it had a supply of the right kinds of atoms, a source of energy and a set of instructions for exactly what to build. We could work out these detailed instructions with a computer and then radio them to the nanobot. If the nanobot could really build anything, it could certainly build another copy of itself. It could therefore self-replicate, much as biological cells do. After a while, we'd have a second nanobot and, after a little more time, four, then eight, then 16 and so on.

For fun, suppose that each nanobot consisted of a billion atoms (10^sup 9^ atoms) in some incredibly elaborate structure. If these nanobots could be assembled at the full billion-atoms-per-second rate imagined earlier, it would take only one second for each nanobot to make a copy of itself. The new nanobot clone would then be "turned on" so that it could start its own reproduction. After 60 seconds of this furious cloning, there would be 2^sup 60^ nanobots, which is the incredibly large number of 1 x 10^sub 18^, or a billion billion. This massive army of nanobots would produce 30 grams of a product in 0.6 millisecond, or 50 kilograms per second. Now we're talking about something very big indeed!

Nanobots in general may not be terribly interesting as a way of making prodigious amounts of things, but self-- replicating nanobots are really interesting. If they are feasible, then the notion of a machine that can build anything from a CD player to a skyscraper in a remarkably short time doesn't seem so far-fetched.

But these self-replicating nanobots can also be quite scary. Who will control them? How do we know that some scienlist or computer hacker won't design one that is truly autonomous, carrying a complete set of instructions for itself? How do we know that these nanobots won't mutate and that some of these mutants won't achieve the ability, like cancer cells, to disregard any signals that would otherwise trigger self-destruction? How could we stop them once they reached this malignant state? Self-replicating nanobots would be the equivalent of a new parasitic life-form, and there might be no way to keep them from expanding indefinitely until everything on earth became an undifferentiated mass of gray goo.

Still more frightening, they would by either design or random mutation develop the ability to communicate with one another. Maybe they would form groups, constituting a primitive nervous system. Perhaps they would really become "alive" by any definition of that term. Then, in the memorable words of Bill Joy, the chief scientist at Sun Microsystems and someone who has worried in print about the societal implications of proliferating nanobots, the future simply would not need us.

But how realistic is this notion of a self-replicating nanobot? Let's think about it. Atoms are tiny and move in a defined and circumscribed way-a chemist would say that they move so as to minimize the free energy of their local surroundings. The electronic "glue" that sticks them to one another is not local to each bond but rather is sensitive to the exact position and identity of all the atoms in the near vicinity. So when the nanomanipulator arm of our nanobot picks up an atom and goes to insert it in the desired place, it has a fundamental problem. It also has to somehow control not only this new atom but all the existing atoms in the region. No problem, you say: our nanobot will have an additional manipulator arm for each one of these atoms. Then it would have complete control of all the goings-on that occur at the reaction site.

But remember, this region where the chemistry is to be controlled by the nanobot is very, very small-about one nanometer on a side. That constraint leads to at least two basic difficulties. I call one the fat fingers problem and the other the sticky fingers problem. Because the fingers of a manipulator arm must themselves be made out of atoms, they have a certain irreducible size. There just isn't enough room in the nanometer-size reaction region to accomodate all the fingers of all the manipulators necessary to have complete control of the chemistry. In a famous 1959 talk that has inspired nanotechnologists everywhere, Nobel physicist Richard Feynman memorably noted, "There's plenty of room at the bottom." But there's not that much room.

Manipulator fingers on the hypothetical self-replicating nanobot are not only too fat; they are also too sticky: the atoms of the manipulator hands will adhere to the atom that is being moved. So it will often be impossible to release this minuscule building block in precisely the right spot.

Both these problems are fundamental, and neither can be avoided. Self-- replicating, mechanical nanobots are simply not possible in our world. To put every atom in its place-the vision articulated by some nanotechnologists-- would require magic fingers. Such a nanobot will never become more than a futurist's daydream.

Chemistry is subtle indeed. You don't make a girl and a boy fall in love by pushing them together (although this is often a step in the right direction). Like the dance of love, chemistry is a waltz with its own step-slide-step in three-quarter time. Wishing that a waltz were a merengue-or that we could set down each atom in just the right place-doesn't make it so.

Corwin

Smalley's arguments are interesting, but he frequently fails to take basic issues into account when he makes a point. (Yes, a nanobot would take 19 million years to make a mole of a substance.... he neglects to mention that we aren't talking about using a single nanite to do this... we're talking about several billion doing it simultaneously.) He also tends to have a chronic case of tunnel vision. He looks at things as a chemist, and frequently ONLY as a chemist. He's a brilliant man, but sometimes he has a bit of a problem stepping back and seeing the big picture.

alek0

He did mention several billions of them.

"This massive army of nanobots would produce 30 grams of a product in 0.6 millisecond, or 50 kilograms per second. Now we're talking about something very big indeed! "

You have missed the point. I haven't seen yet any viable solution for "sticky fingers" problem, "fat fingers" problem, as well as programming and control, and the energy source. Do you have anything else to offer to the objections raised other than "tunnel vision" and "doesn't see the big picture"?

Corwin

There are several possible solutions, as to which will work the best, we don't know yet. When it comes down to it brute force can accomplish a lot. So can simply allowing chemistry to do it's thing. (If the tips of your manipulator arm are less reactive, or ideally even non-reactive with the element you're moving, the problem more or less goes away. All you have to do is move the, let's say carbon atom within a given distance of, let's say two oxygen atoms. The attraction of the oxygen atoms will be greater than that of the maniuplator arm.) This is just one example... using self assembly instead of actual nanites is another. The recent work that Lucent and HP have been doing with molecular wiring is a good example. Instead of moving molecules into place, they set up the structure they want and let the molecules move into the right configuration on their own, using basic laws of chemistry... (well, this molecule will be attracted to this one, and so on down the line...) It's complex, but they're getting promising results with it.

So we have at least one person that has some objection other than 'cuz I says so.' But let's step back. There's a larger issue here than molecular manipulation. (Which I don't discount. You'll notice I don't claim that all of these things will be EASY, just possible. If they were easy we'd likely already be doing them.)

Nanotech offers virtual immortality, an elimination of disease, an end to scarcity, improvements in areas like space travel, electronics, medicine, it can influence virtually every part of human life really. Assuming we can make it work, (and the evidence suggests this is possible, given time, effort and study) why couldn't we, for example, live for thousands of years, or even longer? Most of the 'cuz I says so' arguments tend to avoid any technical discussion and start in with 'axioms' that are really anything but. (People can't live forever. Why not? Because people can't live forever stupid and you're evil for even thinking you can try!!!!!!)

Nat

Feynman famously said that there is "There's plenty of room at the bottom" meaning that you could do a lot with the very small. He was right to a degree (probably to the degree he intended) and the advances in semiconductor technology are case in point. The Nantech advocates, however, have jumped all over this idea hoping to create the ultimate technology that they claim can do virtually anything, yet they never explain how they can get around certain substantial problems.

First, there is not infinite room "down there;" you can't just make things smaller and smaller - eventually you hit a fundamental limit. You can't make the fingers or arms of the nanobot any smaller than simple molecules - and the simpler you make the molecule on the arm, the less specificity you can assign to it. This isn't a question of technology - this relates to fundamental limits of the Universe.

Further, no matter what type of machine you are building (chemical, biological or mechanical), to do work, it must use energy. No one seems to have even begun to answer about how these little buggers would fuel themselves without burning up their host (in medical applications - the most often talked about use of nanotechnology).

Science fiction writers and "futurists" love this concept and dream of all types of amazing achievements using it, but reality seems to be against most of the truly amazing applications.

Cheers

Nat

I think you are jumping the gun here. You haven't addressed the fat fingers problem or the energy/heat concern and you have only partially discussed the sticky fingers problem and you seem to think that the evidence is there for this to technology to do the amazing things dreamed of by science fiction writers?

These problems are real and fundamental. Maybe they can be overcome, but I would say at this points the evidence away from the more fanciful applications at this point.

Corwin

'Fat fingers' isn't nearly as much of an issue as it's made out to be. It's a limiting factor, sure. But we pick things up that are smaller than our own fingers all the time... so do industrial robots. There's a lower limit, of course. (A manipulator arm tipped with a C60 buckyball isn't going to be able to pick up a hydrogen atom with any degree of precision) but that just limits some applications, not the general concept... and can still be worked around. (A manipulator tipped with a helium atom, for example, could pick up and move a hydrogen atom.)

Energy is always a problem. Fortunately there are several natrual sources ready at hand... the problem isn't finding the energy, it's figuring out how to USE it. (For example, ambient thermal energy could be used by the nanite, or the nanite could be designed to use chemical energy sources such as ATP.)

NialScorva

I think you underestimate the thermodynamic problem, Corwin. Consider carving a statue by knocking off 500 mg chungs, then consider how much more energy is required to knock it off 1 gram at a time. It's still going to be easier and more efficient to die-cast or mill parts rather than to use nano-manufacturing. That's the biggest reason I see for nanotech not being in every home-- there's tons of methods that are easier and more efficient at doing the exact same task. My guess is that nanotech will only be useful in tasks where the number of MEMs is measured in the hundreds or thousands.

Corwin

Of course there will be tasks that are easier to do traditionally. There will always be tasks that don't require the sort of precision or automation that nanotech offers.

I don't NEED a car to go to the grocery store either. I've still got functioning legs. To go downtown, however.......