Follow Slashdot blog updates by subscribing to our blog RSS feed

 


Forgot your password?
Close
typodupeerror
×
Technology

The Evolution of Nanomachinery 79

Posted by michael from the grey-ooze dept.
csy writes: "Harvard's George Whitesides has a wonderful article on Nanomachinery in this month's issue of Scientific American. He casts doubts on the Drexlerian vision of mechanical assemblers, and argues that biology and chemistry, rather than mechanical engineering point to the answers in the quest for nanomachines."
This discussion has been archived. No new comments can be posted.

The Evolution of Nanomachinery More

The Evolution of Nanomachinery

Comments Filter:

  • Even smaller scale though? (Score:5, Informative)

    by apsmith ( 17989 ) on Saturday August 18, 2001 @09:20AM (#2171966) Homepage
    The article talks about the machinery of cells as an example of existing nano-machinery on which we should base the development of artificial nano-machines - but the proteins and other bio-molecules in a cell are actually pretty large compared to some of the things we can do even now with STM microscope tips and carbon nanotubes. Even the smallest virus is 0.05 microns across, and we're already regularly making semiconductor components on that scale. Admittedly the virus has some complex internal structure. But biology uses a very limited set of chemical elements (mainly C, H, O, N) and I think one of the main ideas with nano-machines was that there's no need to restrict yourself to the limited set of things used in biology...
    • Have you ever seen an STM? I think it is rather blind to claim that only the size of the tips matter.

    • You are correct, that biology limits itself to C, H, O, and N, and that Nanotechnology seeks to use the rest of the perioidic table.
      However, at the nanometer scale, atom and molecular bonding begin to really dominate, and therefore, architechures at that scale are predetermined by the chemistry of the atoms. The reason that biology uses C, H, O, and N are two fold. First is that they are abundant, but more importantly, its easier to make a wide range of chemical structures with these elements. Ever wonder why there is not a full Silicon version of bezene? Its becasue Silicon does not want to be forced into that sort of molecular structure, and for that matter, neither does most of the perioidic table. Its considered a major achievement in chemistry to make structures seen in nature with elements from the rest of the periodic table - and it takes a HUGE amount of effort to obtain these materials. To basically sumarize, yes, current efforts in Nanotechnology are to make biological and chemical analogs of things found in nature with elements not commonly used in nature. However, the laws of chemistry that govern these elements will pre-determine what we can and can't do. So by copying nature, we have a better chance of success for creating real nanotechnology, and eventually, picotechnology.
      • If you did a small amount of investigation you would know that the atomic composition of currently designed nanoparts (here [imm.org]) is mostly C followed by H/N/O and to a much lesser extent S/Si/P/F. With the possible exception of F (diatoms in the ocean use Si) all of these elements are used by Nature. So the statement that "nanotechnology seeks to use the rest of the periodic table" is incorrect. A better statement would be that "nanotechnology seeks to explore the assembly of atoms in ways that, to date, have been unexplored by nature".

        Current efforts in nanotechnology are not directed towards making "analogs" of things "found in nature with elements not commonly used in nature". Current efforts are directed towards using the laws of physics and chemistry to explore regions of the phase space for atomic structures that are by and large unexplored by nature. Zyvex [zyvex.com] wants to assemble diamondoid materials -- that isn't a different element, its a different way of putting carbon atoms together than that commonly used by nature. Nature primarily assembles polymers (DNA, RNA & Proteins) which involve creating 2 covalent bonds -- molecular nanotechnology seeks to create more rigid, stronger materials by controlling the creation of 3-4 covalent bonds.

        The development of nanotechnology is likely because of the "existance proofs" provided by nature. The development of "picotechnology" is highly speculative, as documented by Hans Moravec in Harvard Doesn't Publish Science Fiction [aeiveos.com].

        It would be nice if people really knew something about a topic before they commented on it.

    • But biology uses a very limited set of chemical elements (mainly C, H, O, N)

      Bulk biology uses a limited set of chemical elements. However the most interesting bits, the enzymes use a far more diverse set of elements including iron, nickel, copper, phosphorus, zinc, and so on at their active center. Most metals have significant biological activity as well. If they didn't things like lead, arsenic and cadmium would be innocuous rather than poisons.

    • If those nano-machines are going to be powered from outside, that's all fine. But that's a far cry from self-sustaining, self-replicating assemblers. As the article explains, breaking the chemical bonds of the atoms you'd use as a part for replica requires some energy. Where do you get it from? When you build the replica, does it automagically contain a battery? Or do you suggest connecting it to some nano electric outlet nearby?


      The only solution is to use energy from the environment, either chemical energy (then nanoassemblers only work when submerged in ATP molecules or something) or photons (by photosynthesis). Any other form of energy supply would be destructible. These two supply strategies roughly correspond (suprise!) to animal and plant life, respectively.


      • Where do you get the energy? How about getting the
        energy to break bonds from the energy released from the similar number of bonds you create?

        You'd near some energy to start up but after that
        you'd just need to give the assembers enough to
        make up for waste energy lost.
      • You can get the energy by running glucose through a nano-fuel cell. You can also carry an onboard diamond pellets that you burn... (C + O2 = CO2 + free energy) You could have diamondoid tanks containing compressed hydrogen at 1000 atm (2H2 + O2 = 2H2O + free energy). Finally, and I like this one the best you can have nanobot nuclear reactors powered by Gandolinium 148 (See Nanomedicine, pg 158).

        There are ways to provide external power as well, you can lie on an ultrasonic couch and the Nanobots can harvest the ultrasonic energy it supplies. Nanomedicine was published in Oct. 1999, so Whitesides would have had plenty of time to investigate this if he had really wanted to give the topic fair treatment.

    • Not much really.

      I have long maintained that the first product that was mass produced was not an automobile as in Henry Ford, rather it was alcohol as in beer and wine.

      That's nano/biotech at it's best.

      More recently we have discovered fungi that make antibacterial materials. Now we can actually do open body surgery and have some real expectation to recover from the wounds.

      Even more recently we have engineered bacteria to make insulin. We have big tubs of these nano-bio-machines. Each one is a tiny self replicating insulin synthesis plant.

      What's the big huff about nano/bio/whatever tech? It's been around for millenia! Only the words used when talking about it are different.

    • C, H, O and N are not as limited as they seem. What's more, bio systems are more than happy to branch into more exotic chemicals when needed, after all, Hemoglobin uses four Fe atoms to do the voodoo that it does so well.

      One of the most interesting things that we are beginning to learn about bio nanomachines is that they tend to rely on their shape more than their bonding properties (except for stuff like hemoglobin, where the bonding properties are necessary, and then they tend to use just about anything). Shape is totatlly doable with a small set of building blocks. So far, we really don't need to second-guess the old pro. If we can do what bio does, then we're already there. Think about the fact that biology has covered the earth in solar panels. If we could find a way of using ATP to power our tools, then we could just pave the streets with chlorophyll and we wouldn't need power plants.
      Wasn't looking to go with that 'plant' pun there.

  • does makes sense (Score:3, Insightful)

    by boaworm ( 180781 ) <boaworm@gmail.com> on Saturday August 18, 2001 @09:26AM (#2171977) Homepage Journal
    One of the things you often hear when talking about open-source software is "dont reinvent the wheel", ie dont do anything which is already done. Late research has found these stem cells, and if we can use them to manipulate the human body, why try to build some nanorobot in metal ? It is already there, we just have to find out exactly how it works.


    The possibilities, if we just can figure them out, are enormous. You can create any kind of human cell out of those, the genetic code is all in there.



    Bio-informatics is probably just in the very beginning of something huge. Once we gain full understanding of the human body we have the code to life itself. It sure is a thrilling thought :-)

    • Comment removed based on user account deletion
    • You're exactly right

      You can take your analogy even further. Using nanoscale tech to build machines is like counting to a billion by ones.
      For those out there that think smaller is better, think of a protien or a cell as a function, a complex array of int float, double etc



      So I can make a protein stand for an complex number, and a nanomolecule can only stand for a constant and non variable int.



      Surely, it makes more sense to have a whole "class" of wheels, than to "redefine" the wheel every time we want to use it.



      A protien is just a bunch of nanomolecules with a bunch of hard coded instructions. Which makes building machines easier, faster, and more efficient. Why have a billion nanos that form a wheel, when you can add sulphate bonds to a protein and allow it conform to a wheel structure. On the same protien, add hydrogen bonds and watch it form pulley. It may not look as pretty as a regular machine, but it'll do the job just the same and most likely, better.



  • If ever there is any nano inspiration, then there would be the combination of Neumann Machines with Nanotechnology.

  • Let's build us a microscopic wheel and put an Amoeba on it.

  • Fab Method (Score:2, Interesting)

    by 1alpha7 ( 192745 )

    If they are to pick up atoms with any dexterity, they should be smaller than the atoms. But the jaws must be built of atoms and are thus larger than the atom they must pick and place.

    Atoms, especially carbon atoms, bond strongly to their neighbors.

    I really don't care if the "jaws" are Drexlerian or biochemical. As long as the damn thing works, we're golden. An assembler that is a mix of mechanical and chemical, or any other approach is still a successful assembler. As for his concern of how to make a machine self-replicating, we don't need the assembler machine to be tiny, just to make tiny stuff. Room for template storage is easy.

    1Alpha7

  • It's not a zero sum game is it? (Score:3, Insightful)

    by ahfoo ( 223186 ) on Saturday August 18, 2001 @09:48AM (#2172010) Journal
    There's room for lots of approaches, I'm sure.
    I felt the author was a bit disingenuous with this quote:
    "A little submarine that was to be a hunter-killer for cancer cells would have to carry on board a little diagnostic laboratory, and because that laboratory would require sampling devices and reagents and reaction chambers and analytical devices, it would cease to be little."
    This is clearly a mix of humor and a rhetorical stab at the nanotech research community. That's fine for a popular magazine like SciAm, but it's not a serious analytical point. We'd be kidding ourselves to pretend that the only possible techniques for identifying cancer cells when parked before them on the nanoscale would require lab reagents and little miniature lab assistants in white coats drawn by Gary Larson.
    Of course SciAm has always been a popular publication masquerading as a scholarly journal and evocative claims have long been the stock in trade.
    Good one, I guess.
    • You're right - but I found the claims made in the article refreshing, as they point out that the chemistry of the elements dictates what one can and cannot do at the nanoscale, not just our limits in fabrication and machining techniques. In a way, hunter-killer cells currently exist, in the form of white blood cells. The antibodies created by chemistry or the human immune system (biochemistry) are the analytical laboratory for detecting viruses, bacteria, and maybe even cancer cells. The main reason that white blood cells don't wipe out cancer for us is that they do not have any antibodies to use to differentiate normal cells from cancer cells. Basically a cancer cell is a normally useful living cell, no different than its neighbors. What makes it different is that its reproductive programming has been set to permament high speed, rather than normal growth mode. So if you want to put this in computer terms, Cancer is basically where there is a bug in the programming (DNA), and the program is set to repeat itself and absorb more and more resources while it does, thus growing in size. If a white blood cell could be programmed (through the use of antibodies or some other genetic programming), it will recognize cancer cells from normal cells and wipe them out for us. Nanotechnology, and all of its potential applications, will be dictated by chemistry and biochemistry, because at the nanoscale, those are the limiting factors for nanoscale design. If the atoms do not want to be in a particular arrangement because it is energetically and entropically disfavored, they will not be in that arrangement. This is why nanotechnologists look to nature, because in nature there are examples of nanotechnology that is energetically and entropically favored, and therefore, it can be built.
      • As I pointed out in another comment, Drexler pointed out these idea that one could look to Nature for micromachines as much as 20 years ago.

        I agree that one can use antibodies as an analogy for a diagnostic laboratory. This is how I think of the "tips" of the molecular sorters that are discussed in Nanosystems and Nanomedicine. However I must disagree with the idea that our white blood cells do not wipe out cancer. This was an unresolved question when I took immunology 8+ years ago. I strongly suspect in people that have the right combination of MHC alleles that they can display the fragments of mutated proteins such as p53 (mutated in over half of all cancers). For some individuals their immune system may quite be quite able to recognize and kill those cells. But those dastardly cancer cells are tricky. It has been shown that cancer cells will mutate to the point where they display a protein known as the FAS ligand which they can use to instruct the immune system cells to commit suicide (apoptosis).

        To get around this you will really need much more sensing capability and programmability packed into the nanobots so they can learn to recognize the many disguises the cancer cells may adopt. Freitas has proposed in Nanomedicine that the nanobots may actually be able to measure the milli-kelvin temperature differences one would find around the more active cancer cells. They would also be able to detect the low oxygen conditions that exist in small tumors before they active the angiogenisis genes to get an increased blood supply. There are lots of things we will be able to do once we have the technology.

    • Of course SciAm has always been a popular publication masquerading as a scholarly journal...


      See Foresight vs. Scientific American [foresight.org] -- they've been adversaries of nanotechnology for many years.

  • There are some interesting ideas here, but I wish they had been presented as such, rather than as evidence that nanomachines won't work. Nanomachines can't self-replicate by placing individual atoms with tiny pincers, therefore self-replication is impossible. Nanosubmarines 100nm in size would be too small to steer, therefore nanosubs are impossible. Jeepers, use a little imagination.

    I think his claim that the molecular assembler is "less the solution of a problem than the hope for a miracle" will seem quaint in time to come.
  • First of all thank you for the article.

    Second, think about it. The ideas in this article bring some debates. We as humans are always trying to make computers life-like, trying to induce a sense of self with in. We want the computer to make a choice with out the guidance of a complicated algorithm. If we do use the approach suggested in this article than these nano-machine will eventually develop a conscious!

    Do we wanna play this game?
  • I'm so concerned whether they are made of organics or metal (I used to teach chemistry and physics), but I have some work for them to do right away.

    Memo to future small robot type guys:

    1) Clear out any hardening of the arteries that I have already developed (I'm 50 years old)

    2) I hate flossing--could you do that stuff by launching from my mouthwash?

    3) About that pain in my lower back? Could you head down there an do a diagnosis? I would hate to let a bone cancer get a headstart. Oh, if there is a tumor or something bad, please take care of that while you're there.

    4) Say, Viagra work but it is very expensive. Could a bunch of you tools help my tool--every once and a while? But please give me warning!

    5) Hey, I know that this is not so important, but the Q-Tip package says don't use them in the ear canal. Like many others I have been feeling guilty about this for years. Could you guys...you know...that waxy stuff?

    6) OK up here in the macro world we have these utilities called McAffee and Norton that look for problems and then if I give the OK, they fix what they can. Well, can we work out something like this, only on your level.

    7) Gee. I almost forgot being a diabetic. You guys need to fix that first.

    8) And taking that anti-depression medicine Zoloft..could you do something in my brain. BUT PLEASE BE CAREFUL THERE GUYS!

    Thanks,
    Brent,
    Your Commander-in-Chief

  • Moreon NANOs (Score:2, Informative)

    by cdtoad ( 14065 )
    A few months back Machine Design magazine did a good article on the coming of Nano technology. It's on their website at http://www.machinedesign2.com/turnstyle.php?ID=100 0 [machinedesign2.com]... and pretty much gives a good 50,000 foot view. Only draw back is that it's in PDF again.

  • This article is just plain wrong. The Drexlerian vision is based on Wet Nanotechnology (meaning biology and chemistry). Eric Drexler is very much against the notion of Dry Nanotechnology - in fact, most of his blueprints for 'nanoparts' are based on proteins.

    What is it with all these morons at the universities trying so hard to discredit Drexler? Are they jealous that they didn't think of it first? It sounds like it.
    • Eric Drexler is very much against the notion of Dry Nanotechnology


      Ah, are you sure? I've never heard him say as much, and he seems to get along fine with people who are actively working on dry nanotech. How did you come to the claim that he is "very much against the notion"?


      -- MarkusQ

  • Existance proofs (Score:5, Interesting)

    by steveha ( 103154 ) on Saturday August 18, 2001 @10:11AM (#2172054) Homepage
    This article contained a lot of straw-man attacks against Drexler's ideas. But it missed the point. K. Eric Drexler never said that his vision of little mechanical machines is the only way to go about things.

    For example, Drexler focused on mechanical computers, with little rods moving back and forth. Does he think nanoscale quantum computers, driven by electricity, can never work? No, but for his book he wanted to focus on things he could be sure would work. Because nature includes little machines, he was sure you could use little machines to build things like tiny computers.

    His first book, Engines of Creation [foresight.org], is pretty much about existance proofs. He figures you can probably make an assembler with just 150 million atoms, so then he assumes it will take a billion atoms (just to be on the safe side) for the rest of the discussion.

    And in his discussion of how we will get these magic assemblers, he said that one possible route was biological: use tailored cells to make new cells that are closer to what we want, and iterate. He isn't ignoring biology, or reality.

    The article is weak. Read Drexler's book instead; it's online so you can read it now for free.

    Engines of Creation [foresight.org]

    steveha

    • Okay, I'll bite at this one and take a read at Drexler's book.
      I don't think you should look at this as an attack directly on Drexler's work, but more of a bit of realism on the limits of nanotechnology. In a way, chemists (though probably not thinking about it this way) have been practicing nanotechnology for hundreds of years. They build nanoscale strcutures all the time by simply reacting chemicals together in a flask. It doesn't sound very glamourous, but they are building nanoscale structures, as all those atoms are in precise locations, yielding a chemical structure which has certain properties and effects. The plastics industry, through the use of catalysts, has had molecular assemblers in place for 50 years. To be specific, they can take a small 3 carbon molecule (propylene) and use the catalysts to assemble a long-chain polymer one piece at a time, with atomistic repeatable control, to produce long-chain molecules with millions of carbon atoms present.
      What scientists are saying when they comment against Drexler's views is that he doesn't go into the chemical specifics, and nor does he acknowledge that versions of his devices currently exist in nature or in chemical reactions. They just put things together with chemcial bond-forming reactions rather than single atom "manipulators".
      • What scientists are saying when they comment against Drexler's views is that he doesn't go into the chemical specifics, and nor does he acknowledge that versions of his devices currently exist in nature or in chemical reactions.

        If that's true, then it's out of pure ignorance of his views and his work. He very much does go into chemical specifics in Nanosystems, his technical work (as opposed to Engines of Creations which is meant for popular consumption), and one of his favorite replies to denials of the possibility of nanotechnology goes something like "We are nanotechnology." So much for denying that versions of these devices exist in nature.

        In the very introduction of Nanosystems he compares and contrasts normal lab chemistry (solution-phase chemistry) with nanotechnological construction (mechanosynthetic chemistry).
      • I don't think you should look at this as an attack directly on Drexler's work, but more of a bit of realism on the limits of nanotechnology. In a way, chemists (though probably not thinking about it this way) have been practicing nanotechnology for hundreds of years.


        I think that's the root of most of the missconceptions; they haven't been practicing nanotechnology for hundreds of years, they have been putting things in tubes and shaking them. They have learned an awful lot in the process, but they have also picked up some amazing blind spots. The fact that "you can't assemble something that complex by putting the parts in a box and shaking" does not mean "you can't assemble something that complex, and it wouldn't work if you could."


        In a sense, their knowledge works against them.


        -- MarkusQ

      • Drexler does go into chemical specifics. Chapter 8, pgs 191-249 of Nanosystems [foresight.org] is all about such specifics. He also points out the fact that devices in Nature can be used as existance proofs for these ideas. From is 1981 paper [imm.org], "The existence of this range of components in nature indicates that power-driven mechanical systems can be constructed on a molecular scale."

        There are two problems the chemists have. First, they haven't read the material. Second, Drexler is proposing to precisely assemble millions to billions of atoms and the chemists think that is infeasible. That is why programmers can accept nanotechnology to a greater degree than chemists -- manipulating a million or a billion "bits" is something they regularly have to deal with. For chemists the idea is nightmare.

        I'd urge readers to educate themselves with regard to the material before they comment on it. If we have to spend all of our time attempting to erase malformed memes we will never get a chance to work on developing new ones.

        • As a chemist, I can say billions or even trillions of atoms don't bother me at all. I work with "Moles" (6.022X10^23) of molecules and atoms every day.

          I will take some of this criticism and read Drexler's work in detail. Perhaps I am missing something that I'm not seeing from reviews of his work. Still - we have chemical analogs of his devices today, we (and nature too for that matter) cannot place a single atom (unless its completely inert and near absolute zero) exactly where we want it to go if it does not want to be there due to unfavorable chemical bonding. If I had to guess why Chemists pooh-pooh Drexler's ideas, its that his proposed nanoassemblers are quite slow compared to chemical reactions. His devices could make what - millions of sub-devices or parts per day? In one chemical reaction, well designed and with use of catalysts and substrates which force the chemicals into one and only one configuration, I can produce far more in the same amount of time (remember, I can operate on molar scales with ease), even if my chemical yield for the reaction sucks.

          What I think the real advantage of Drexler's views may be is getting conventional science to think differently about how it designs materials at the nanoscale, and how those nanoscale structures can be further manipulated in ways which chemistry cannot do (nanobots for example).

          • I'll grant the point that chemists work with lots of molecules. That doesn't however mean that you can approach Drexler's vision because you potentially have a big problem with chemical reactions yielding nonspecific products.

            Drexler's vision is very clear, you want to get almost all of the atoms precisely positioned and bonded in 3D space developing machine parts on the nanoscale from 50-500 nm. Thats millions to billions of atoms. The largest specific structures chemists build contain only a few hundred atoms. And getting there isn't easy -- consider it took around 30 years from the discovery of Taxol until we figured out how to synthesize it.

            And assemblers need not be slow. See this paper by JoSH Hall. A properly designed nanotech assembly line can reproduce its weight in nanoparts every millisecond! However its good that you are interested in this. I'll be publishing a paper soon that will I hope will shorten the path and decrease the cost to achieving real molecular nanotechnology. And it uses organic chemistry synthesis as its foundation.

            • Chemistry is all about getting specific products, otherwise when I mix chemical A and B, I'll get products C to Z in one big mess. By chemical synthetic design, one can get very specific products.

              As for small structures containing only hundreds of atoms, I'll agree on that point, although polymers are an exception to that rule, as one can synthesize polymers containing millions of atoms with speicific atom placement (thanks to catalysis). One of my concerns is that the nanoassembler and nanoparts must have some sort of weak-bonding interactions to work. If the atoms (or molecular chunks) have any more than van der waal (weak bonding) interactions, I feel that the whole system would get stuck and gunked up as more and more atoms started bonding to the nanoassembler, rather than to the desired part. I would think that to avoid this you would need to have a very low temperature environment, or your design would have to be very precise. And what happens when a cosmic ray hits the nanoassembler and destroys the whole thing? Obviously, you'll need a huge series of nanoassemblers, but how do you separate those parts damaged by cosmic rays from those which are not damaged?

              Regardless, you've given me a lot to think about, and I need to read up more on what has been proposed and see what I think then. My views are based on reviews of the work, rather than sitting down and reading the original and using my chemical expertise to form my own opinion. Based on what you and others have posted, I need to do more background research. I suspect the real solution of nanotechnology will be much different than what any of us understand, as we continue to learn new phenomena which govern the nanoscale. I'll keep an eye out for your article, where are you going to publish it?

    • And if Engines of Creation leaves you hungry for technical meat, try Nanosystems [amazon.com].

      -- MarkusQ

    • I'm not going to go through EOC to see if the quote is correct. My guess is that 150 million atoms is more like the requirements for an assembler system or perhaps even a self-replicating system. Assembler arms as proposed by Drexler (in Nanosystems) or Merkle (the Stewart platform based assembler) have been estimated to require 4,000,000 and 3,000,000 atoms respectively.

      If we are going to discuss nanotechnology we need to be precise. :-)

      • I'm not going to go through EOC to see if the quote is correct.

        No need; it's in Chapter 4 [foresight.org]. Just follow that link, and search for the word "billion", and read the paragraph under your cursor.

        My guess is that 150 million atoms is more like the requirements for an assembler system or perhaps even a self-replicating system.

        150 million is for a general-purpose assembler capable of self-replication. And then he rounds up to a billion just to add margin for error.

        steveha

  • Some Basic Rules (Score:3, Insightful)

    by the eric conspiracy ( 20178 ) on Saturday August 18, 2001 @10:40AM (#2172099)
    Nanomachines as scaled down miniatures of human scale machines is clearly very unlikely to materialize, for exactly the same scaling reasons that prevent us from having human size single cell organisms. Fundamental relationships between mass, surface area and linear dimensions are inescapable. These relationships govern the nature of physical structures of all sizes. Clearly at nanoscale the balance has shifted from a dominance of bulk properties to surface properties that are primarily chemical in nature. Anyone trying to translate an instrumentaility to nanoscale from human scale will fail miserably if they fail to account for the basic physics of scale.

    Innovative nanomachines that make use of atomic scale forces are another thing altogether. As Whitesides correctly points out, this is the realm of chemistry and biology, not the mechanics of bulk materials.

    • While all these rods and gears and things may sound like a silly application of macro-scale approaches to micro-scale systems, it actually is all based on atomic-scale forces.

      We know from experiments with various tiny-finger-type microscopes that you really can push around atoms as if they were little beach balls, and that bucky tubes really do act like fairly stiff, yet flexible, rods. They really can act mechanically on each other in reliable, predictable ways.

      Nanosystems uses these interactions to argue for the possibility of nanotechnology because they are simple and easy to understand. Every argument is reinforced with large fudge-factors and cautious assumptions (for example, it is assumed that any machine will become non-functional or malfunction if a single atom is out of place).

      Nobody is qualified to criticise Drexler's work until they've actually read it, and Nanosystems is the real meat of his work. It's also a great book if you'd like to learn more about any of chemistry, mechanical engineering, physics, or computer science, because of the way it ties them all together. The math is heavy going, but in its own way it is every bit as worthwhile to dig through as Knuth's TAoCP.
    • The Bandersnatch is a single-celled organism, and it's much larger than a human being :)
  • I've been waiting for someone else to say there is never going to be a "gray goo" problem. What a refreshing change from the "nanotech is coming, we need to regulate it before it eats everything" nonsense.

  • The author is missing the point (Score:4, Informative)

    by MarkusQ ( 450076 ) on Saturday August 18, 2001 @11:31AM (#2172229) Journal
    This isn't a rebuttal of Drexler so much as an ignorant dismissal. The "problems" he finds with eutactic nanotech all seem to fall into a few buckets:


    1) Assuming that everything is like silicon (e.g., the MEMS/stiction arguments). This is like arguing that skyscrapers are impossible based on the properties of beeswax.


    2) Assuming that everything is like wet chemistry (e.g. all the comparisons to biology). He even tries to draw a dichotomy between these two.


    3) General bad logic. The self-replication argument, for example, flows as follows: We don't know how to make anything that self-replicates at present; cells replicate; they do it by assembling things in a linear sequence, rather than 3d; therefore this is a serious problem for nanotech. Not only are all of these steps factually suspect (enzyme structure, for example, is very much a 3D proposition), they don't logically lead to the "conclusion".


    4) Strawman arguments; saying things like "Machining and welding do not have counterparts at nanometer sizes" when no one claimed they did, or "There are no electric sockets at the nanoscale" when no one claimed there were.


    This isn't exacty an objective or even rational rebuttal to Nanosystems [amazon.com] or any of Drexler's other work; instead it seems to be an attempt at persuasion based on the author's knowledge of his own field and ignoring what Drexler actualy wrote.


    -- MarkusQ

    • Well, to address your points:

      1) The problem with nanotech is the assumption that familiar macroscopic concepts (gears, pulleys, "assemblers", what have you) can be easily translated to nanoscopic dimensions. In general, they can't because the physical and chemical properties at that scale are different. Stiction is just one specific example of that.

      2) Everything is exactly like what it is. Some things are somewhat like things they aren't. For the physical world, the degree of "somewhat" is pretty much independent of convenient or attractive metaphors (unless you believe in deconstructionism). But when you're trying to communicate nonobvious concepts to nonspecialists, you're doomed to using everyday language, which can easily be imprecise, vague, or misleading. Some scientists (Sagan comes to mind) seem to be better at this than others, but they also tend to spend more time at it.

      I think it's highly unlikely that Whitesides assumes everything is like chemistry or biology. What he's trying to do is demonstrate that a) we already have nanotechnology (biology) that works so well that we have yet to improve upon it in some of its most mundane aspects, yet we're nowhere near to understanding it, and b) nanotechnology in the sense of nanoscopic analogs of macrosopic mechanical devices is likely to be quite a bit more challenging than common technical sense might suggest.

      3) Enzyme structure is 3D, but that 3D structure is determined by its linear sequence of amino acids. How that 3D structure is achieved continues to be a mystery. It's not random thermal motion causing the strand to move through all physically possible conformations until the proper one is found; it's been shown that to do so for any reasonably sized protien would require time on the order of the lifetime of the universe. In some cases, other (3D) protiens appear to assist in the folding of protiens (in 3D space) but of course those helper protiens had to start somewhere... the old chicken & egg problem. Now, how does this relate to a serious problem for nanotech? Well, the point he left out is how incredibly efficient this process is in biological system in terms of yield, minimal waste products, accuracy & error correction, and total energy budget. Compare the energy used to bond an amino acid onto the end of a growing protien with the energy consumed by an STM tip when it's moving single atoms around.

      Using an STM on a single atom is the equivalent of using an entire skyscraper to skewer & move a pea-sized chunk of beeswax.

      4) See my answer in 2). Same problem. But if weak arguments make you queasy, don't think too deeply while reading Drexler.

      I have two main problems with Drexler's modus operandi. First, none of his work provides any, ANY, practical or useful approaches for attacking the fundamental problems that face nanotechnology. Pretty pictures drawn with molecular modeling software and parroting the jargon of the physical sciences doesn't count. He's a techno-utopian; he can imagine the promised land, but he can't help you get there. There's more useful technical content in Feynman's original "There's Plenty of Room At the Bottom". And Feynman's essay is about a library shelf shorter than Drexler's collected screeds, and a lot more entertaining to read.

      Second, because of his approach--employing attractive but technically directionless metaphors--he is effectively tapping into the collective sci-fi raised consciousness of legions of geeks to--dare I say it?--create a religion. Don't believe me? Look at the other posts to this article; look at the posts to any article on nanotechnology, and their uncritical acceptance of our technological reentrance into the Garden of Eden, Real Soon Now. Look at the idiotic ways that nanotechnology is being promoted (objects copied as freely as MP3s, the end of physical want, resurrection of the cryonics cult popsicles) and look at the idiotic ways it is being opposed (Bill Joy's Chicken Little essay).

      There's real, viable, important progress to be made in nanotechnology. Grandstanders like Drexler aren't helping.
  • Planes are constructed very differently from birds, bats or insects. Evolution can often be a very messy designer, engineers are much cleaner. We can look to nature for ideas, but not neccessarily solutions.

    • Eh? Planes are inferior to birds. Planes stall, birds NEVER do. Planes are limited in maneuverability relative to birds. Moreso when compared to insects.


      It is impractical to make an ornithopter based on a bird or insect model - it doesn't scale well, so engineers had to compromise and accept an inferior-performing model as a means of transport.

      • Eh? Planes are inferior to birds. Planes stall, birds NEVER do.

        What are you talking about? If a hummingbird stops flapping, it will stall. Watch a bird landing sometime; it will deliberately stall -- lose forward momentum -- in order to land safely.

        Planes are limited in maneuverability relative to birds.

        And birds are limited in speed and carrying capacity relative to planes. Planes are not an inferior-performing model, just a different one.

        • To clarify, birds and insects do make use of stalls, yes, but they never crash as a result of stalling. They do not unintentionally stall. When they do stall they are capable of recovering virtually immediately because their lifting surfaces are fully reconfigurable.


          I was simply making a point that it is foolish to assume that engineers are going to do "better" than nature, almost inspite of nature. Nay. They would do better to learn from nature which does most things pretty damn well, and in many cases FAR better than an engineer can dream of doing. There is intense research on housefly flight because they can do what is simply, and seemingly, impossible as a matter of course. No problem for the housefly but a MASSIVE problem for an engineer.


          • To clarify, birds and insects do make use of stalls, yes, but they never crash as a result of stalling. They do not unintentionally stall.

            How, exactly, do you know this? I find your assertion dubious. Are you sure that no sick or tired or hurt bird, perhaps flying in unexpectedly bad weather, has ever accidentally stalled?

            I was simply making a point that it is foolish to assume that engineers are going to do "better" than nature, almost inspite of nature.

            And I was making the point that the example you used doesn't work. In the case of flight, engineers did do "better" than nature, much better. Compared to birds, planes can travel farther without stopping, they can travel higher, they can travel faster, they can travel safely through more extreme weather conditions, and they can carry more. All of those constitute "better".

            What's more, planes probably crash less often than birds do, although that's not really a fair comparison because birds are trying to do harder things than planes attempt. Birds often crash into plate-glass windows; planes rarely do that outside of the occasional Airplane movie. Birds sometimes crash into other things while trying to catch prey or avoid predators. Baby birds often fall out of nests whereas young planes rarely fall off of runways. :-)

            Yes, it's useful to study nature. But it's foolish to assume we can't better meet our needs by improving on nature given that our purposes are different than those natural selection was trying to optimize for when it came up with whatever local maximums we see around us.

    • " Planes are constructed very differently from birds, bats or insects. Evolution can often be a very messy designer, engineers are much cleaner. We can look to nature for ideas, but not neccessarily solutions."


      Other than going very fast (woo!), any flying apparatus ever designed by human engineers is a clumsy inefficient piece of crap compared to a dragonfly, much less a bird or bat. You picked a really hideously poor example.

  • Biologically occurring chemical designs are REALLY very good at what they do. So efficient, that we might be tempted to flat out state that even clever mechanically engineered designs will never approach them. But we should keep in mind that in the "design space" that biological evolution works in not every possible idea is tried or tested. Its just that there is the potential for any combination to be explored.

    Human (or machine) created designs can come up with "new" ways to do things. Some of those designs might be more efficient in certain applications than biological designs. A simple example might be the wheel. The last time I checked no macroscopic organisms on this planet where using wheels to get around. On flat surfaces (deserts/ice/savannas/etc.) wheels are more efficient than legs. It is not that evolution couldn't "invent" wheels, it either hasn't successfully happened before, or the design was discarded in favor of the more functionally flexible limbs we do posses. Check out the GOLEM project [brandeis.edu] for more ideas.

    I do not think that biological designs have the end all say on nanotechnological possibilities. But the science of biology currently has far more small scale tools and vastly greater experience (nature's experience, that is) at its disposal for the development of useful nanotechnology.

  • I think I'm going to open a shop and sell nanobots to new-age medicine types.

    Me: Thanks for the $200, now this pill and the nanobots will do their thing.

    Customer: I can see through this gell-cap, there's nothing in them.

    Me: Ah, that's the beauty.
  • Biology makes excellent nanosystems, and it may well turn out the only way to go. However, I would keep an open mind. Biology has had only a very limited range of materials to work with, and it needs to work under a lot of oddball constraints. The cellular architecture of life, where every unit of life is almost identical and carries a complete blueprint, is convenient for biological systems, but it represents only one of many possible ways of organizing small machinery.

    In fact, physicists have developed a number of techniques that let us manipulate atoms in ways that nature does not use for building complex structures. For example, STMs and lasers allow us to move single atoms. And we have found out that individual atoms and assemblies are much more stable than previously thought.

  • Before the ink was dry on the Whitesides article I had sent a letter sent to the editors of Scientific American [sciam.com]. The Whitesides article contains clear errors as well as misleading statements.

    You may find an expanded copy of my letter to the editors here [aeiveos.com].

    Whitesides is a chemist and while he has made huge contributions to that field, particularly with his nano-imprint lithography, for which he won a Foresight Prize [foresight.org] several years ago, he is not, unfortunately, someone who understands molecular nanotechnology. For that you have to read Drexler's take from the same issue which is here [sciam.com].

    Readers of scientific literature must do "reputation" analysis. Would you trust a life-time COBOL programmer to comment on whether or not your JAVA code was well written or crap? I think not. One should judge the Whitesides article from the same perspective.

    • This is why Fleischman and Pons were ridiculed for Cold Fusion. They weren't physicists.
      Interesingly enough, cold fusion has been reproduced in many labs. So go figure.
      • What may have been reproduced is an excess production of heat in the experiments. I know of no peer reviewed papers that demonstrate that cold fusion really exists. I wouldn't care if Fleischman and Pons were geologists if their proposal didn't violate generally accepted concepts in physics. If you want to violate generally accepted laws you probably need both extraordinary evidence and a good explanation for it.

        The cold fusion debate is different from the nanotechnology debate. Nobody has made a reasonable case (in over 20-40 years) for why it violates any laws of physics. Whitesides gets the whole nanobot-brownian motion discussion wrong because he picks the wrong size for nanobots. They aren't 100 nm in size they are much closer to 1 micron in size. He also clearly hasn't read Nanomedicine, which discusses a number of means for nanobots to navigate. For his argument to hold water he would have to negate all of the navigational strategies discussed there. He is simply uninformed.

  • They're called muscle fibers.

    I lerned this when m ywife took anatomy. our muscels are actually based on a molecular motor/ratchet mechanism.

    Way cool.

Slashdot Top Deals

There is no opinion so absurd that some philosopher will not express it. -- Marcus Tullius Cicero, "Ad familiares"

Close