Those of you with backgrounds or interest in nanotechnology have probably been following the ongoing debate between uber-nano-geeks, K. Eric Drexler and Richard Smalley. This debate has recently spilled out of the rather narrow nanotech community into the larger scientific community with this article in Chemical and Engineering News, a publication of the American Chemical Society. Drexler and Smalley both have impeccable credentials in the the field: Smalley won the 1996 Nobel Prize in Chemistry for the discovery of fullerenes, and Drexler is a well-known theorist with many publications in the field. Both men were featured in Wired 3.08 (August 1995) in a piece on nanotechnology, along with other nanotech luminaries Robert Birge, J. Storrs Hall, and Donald Brenner. (Full disclosure: I worked as research assistant for Don Brenner while I was at North Carolina State.) The debate at hand is the feasibility of what are common termed "molecular assemblers." Read on for a description of the problem, and my opinion as a former nanotechnology researcher.
Lets talk about molecular assemblers first. The term refers to a class of theoretical devices that essentially build things one atom or a molecule at a time. Think of this as a set of little tiny fingers plunking down atoms in precisely the right spots, and you have the basic idea of what is going on. While molecular assemblers are only a small part of the vast field of nanotechnology, they are the most (over)hyped and most feared. The infamous "grey goo" fear, of rampaging nanotech destroying the word, derives from fears of molecular assemblers gone awry.
The vision of tiny fingers moving atoms around is not a terribly accurate picture of the way molecular assemblers work. This is immediately obvious if you think about the problem of trying to pick up a single atom. What do you hold it with? A physical device, like a pincer, could not hold the atom, since it too must be composed of atoms. Instead, most nanotech luminaries speculate that assemblers will work by using fast-changing specific chemistries to grab, move, and align atoms.
Smalley, some time ago, began to lecture on the impossibility of this approach. He carefully and clearly labelled his two problems with assemblers as the "fat fingers" problem and the "sticky fingers" problem. (Scientific American, Sept. 2001) The essence of each argument is that it is impossible to control the chemistry of a reaction by purely mechanical means. The "fat fingers" argument states that for a manipulator to be sufficiently complex to maneuver single atoms, the number of atoms necessary to build it would make the manipulator too large on the chemical scale (in Smalley's definition, 1 cubic nanometer). The "sticky fingers" argument is somewhat simpler, stating merely that there is no way to control the chemistry between the manipulator and the atom being manipulated.
When Smalley advanced this thesis in SciAm, a careful rebuttal was made by Drexler and several others, including J. Storrs Hall. The rebuttal gave specific counterexamples (including one from a research paper by Don Brenner that inspired a followup project that I worked on.) The debate continued outside of forums monitored by the general community until recently. Both Smalley and Drexler have valid objections to the other's arguments. Smalley points out that Drexler's purely mechanical, in vacuo approach is naive in regards to the complexities of chemistry. Drexler points out that current theory and experimental evidence indicate that his approach will work, and that Smalley is misrepresenting the scale of what is required to accomplish nanoscale manipulation of molecules.
After reading the point-counterpoint, and thinking about it in terms of the research I have done in the field, I think I understand the key of this debate. This is still an active field of research, so one of Smalley's objections, that Drexler did not have all the answers for how to control the interactions mechanically, doesn't ring true. If he had the answers, there'd be no need for the research; it would be a development problem. Smalley accurately has identified what Drexler and his colleagues should be researching.
Smalley's statements seem to indicate that he distrusts the theories currently advanced by the computational scientists, both those doing density functional (DF) calculations and molecular dynamics (MD) calculations. It has only been recently that the problem of chaperone molecules and solvents have been approached in a systematic, large scale way. Some of the research in this field is being done by my former colleagues at LSU (now at USC).
However, the results for an "in vacuo" approach have been shown time and again to be reasonably accurate, and as yet, no predictions made from these results have turned out to be invalid. My own doctoral dissertation showed that the relative hardnesses for ceramics can be predicted by computational means.
While Smalley is correct when he says that
Much like you can't make a boy and a girl fall in love with each other simply by pushing them together, you cannot make precise chemistry occur as desired between two molecular objects with simple mechanical motion along a few degrees of freedom in the assembler-fixed frame of reference. Chemistry, like love, is more subtle than that. You need to guide the reactants down a particular reaction coordinate, and this coordinate treads through a many-dimensional hyperspace. ... Chemistry of the complexity, richness, and precision needed to come anywhere close to making a molecular assembler--let alone a self-replicating assembler--cannot be done simply by mushing two molecular objects together.
He neglects the fact that Drexler and his colleagues can accurately predict which of these "hyperspace trajectories" (those of us in the field refer to them as phase-space trajectories) will yield the desired product using MD and DF techniques. This will allow those people trying to construct the assemblers to winnow out unproductive geometries quickly.
Further, Drexler makes an excellent point that he does not expound on, which is that mimicking biological processes may give us unexpected leaps in the construction of assemblers. Ribosomes are almost exactly "what the doctor ordered" in this regard, building proteins according to instructions given by a strand of RNA. Smalley points out that this complicates the chemistry immensely. He is correct, but neglects to point out that by stealing ideas from Nature, we may be able to make it work without necessarily understanding all of the complexities.
My personal feeling is that we will not have to steal much from Nature, other than ideas. While Drexler does not come out and say this, his simplifying approach of trying to work outside of the complex water-solvent chemistry is simply what all physicists learn: the best way to solve a problem is to eliminate unnecessary complexity first. While it may turn out that some of this complexity is unavoidable, the divide-and-conquer strategy will usually beat out a more wholistic approach.
One thing about Smalley's objections puzzles me: in the aforementioned article in Wired magazine, each of the nanotech gurus made date predictions on certain aspects of nanotech. Smalley was the most optimistic about the production of a molecular assembler, predicting that it would be created by the year 2000. Something happened since then to change Smalley's opinion, apparently. My guess is that the "grey goo" hysteria about nanotech, and the desire by certain groups of uneducated, reactionary activists to regulate nanotech out of existence before it ever arrives, has led Smalley to agitate against funding projects that are attackable in this way. By funding more esoteric nanotech research, it might be possible to sneak under the regulators' radar.
This is a very cynical thought on my part, and one I desperately hope is not the case. While science and politics have always been uneasy bedfellows, bringing such Machiavellian tactics into the scientific debate can only cheapen the field.
Posted by brent at 15:34