Is the Revolution Real?

Debating the future of nanotechnology

In his famous 1959 speech, "There's Plenty of Room at the Bottom," physicist Richard Feynman articulated a vision later called 'nanotechnology'. Feynman proposed that mechanical systems (now termed molecular assemblers) could direct chemical reactions, building atomically precise products. This molecular manufacturing process will enable digital control of the structure of matter, revolutionizing areas ranging from military to medical, from environmental to economic. This vision of nanotechnology helped launch the current global surge in research and spending, including the multi-billion dollar U.S. National Nanotechnology Initiative (NNI). Molecular manufacturing has been the focus of Drexler's work, including Nanosystems: Molecular Machinery, Manufacturing, and Computation (Wiley/Interscience, 1992).

In his April 2000 Wired article, Bill Joy voiced concerns about potential abuses of this powerful technology. Fearing the threat this posed to government funding, Prof. Richard Smalley, the leading scientific spokesperson for the U.S. National Nanotechnology Initiative, attempted to dismiss these concerns. He argued in Scientific American (Sept. 2001) that the Feynman vision of nanotechnology is impossible.

In his article, Prof. Smalley declared that "'There's plenty of room at the bottom.' But there's not that much room....To put every atom in its place -- the vision articulated by some nanotechnologists -- would require magic fingers." The debate over molecular assemblers and molecular nanotechnology (MNT) reached a critical juncture in April 2003 when Eric Drexler, through an open letter, challenged Prof. Smalley's misrepresentation. The open letter pointed out that impossible "magic fingers" are unnecessary and had never been part of the original Feynman vision or any subsequent technical proposal. Prof. Smalley, replying to the open letter (Chemical & Engineering News, 1 Dec. 2003), now agrees that assemblers (without impossible "magic fingers") could use something like enzymes or ribosomes as tools for doing precise chemistry.

Prof. Smalley, however, still vehemently rejects the understanding of MNT developed through research over the past decade. In the final part of the C&EN exchange, he mistakenly argues that MNT must use a limited, water-based chemistry, which would exclude the synthesis of structures described in the research literature. His argument contradicts established chemical facts regarding enzymes.

Despite this controversy, the Feynman vision of MNT continues to inspire students and researchers around the world, and the public increasingly expects MNT as part of their future. However, based on false arguments, the U.S. National Nanotechnology Initiative has rejected MNT, thwarting students and crippling research. This is unfortunate, because research in pursuit of MNT offers fruitful areas for scientific discovery and practical application. It is time to reverse this obstructive policy, opening the door to progress toward understanding, developing, and guiding this revolutionary technology.

MECHANOSYNTHETIC REACTIONS Based on quantum chemistry by Walch and Merkle [Nanotechnology, 9, 285 (1998)], to deposit carbon, a device moves a vinylidenecarbene along a barrier-free path to bond to a diamond (100) surface dimer, twists 90 to break a pi bond, and then pulls to cleave the remaining sigma bond.

Isn't this debate just a technical difference of opinion?

The future of nanotechnology is of crucial importance to the future of the world. Research directed toward the Feynman vision points the way to a revolutionary technology based on enormously productive, clean, and precise manufacturing. The potential benefits and risks of MNT for health, arms, the environment, and the economy are huge. If real, they cannot safely be ignored. The Smalley camp, however, had declared MNT impossible and still urges that we ignore it. At stake is both the direction of research and our understanding of where it leads.

Isn't the U.S. government actively pursuing MNT?

The U.S. established the National Nanotechnology Initiative (NNI) in 2000, motivated by the enormous promise of the Feynman vision of MNT. However, actions by the leadership of the NNI show lack of support for MNT research and hostility toward this key goal (while claiming that their unfocused research program will somehow deliver the revolutionary beneits, risk-free). Prof. Smalley, the leading scientific spokesman for the NNI, has persistently misrepresented and ridiculed the fundamental concepts of MNT. Of the two dozen presentations at the NNI's 2003 annual conference, none discussed MNT, and of the billions of dollars of government funding, the amount directly allocated to MNT was nil. The NNI supports work in a wide range of areas, some of which will undoubtedly contribute to the emergence of MNT, but the initiative's leadership ironically now opposes the very vision that inspired young scientists to enter the field.

Why does Prof. Smalley still reject MNT?

As Chemical & Engineering News Deputy Editor-in-Chief Rudy Baum observes, "Smalley's objections to molecular assemblers go beyond the scientific. He believes that speculation about the potential dangers of nanotechnology threatens public support for it."

Prof. Smalley now concedes the central scientific issue (he recognizes that nanomachines can guide chemical reactions without impossible "Smalley fingers"). Nonetheless, his conclusion -- that a broad and powerful MNT is impossible -- has remained fixed and vehement. MNT research envisions highly productive, atomically precise factories able to make, for example, laptop computers. Prof. Smalley, however, persistently portrays it as a "fuzzy-minded nightmare dream" about self-replicating "nanobots" (so-called "gray goo") that could eat the world. Legitimate concerns about artificial self-replicating systems arise in biotechnology, MNT, and other areas, but Prof Smalley fears that concerns about abuse of MNT will threaten government support for the NNI, which funds his research. He seeks to silence these concerns by misrepresenting and attacking MNT. This is ironic, because MNT (unlike biotechnology) need not rely on artificial self-replicating systems.

Ignorance of existing research results may play a large role in Smalley's position. Ralph Merkle, nanotechnology pioneer and Distinguished Professor of Computing at the Georgia Institute of Technology, observes that "Smalley hasn't acknowledged the extensive scientific and technical literature on mechanosynthesis -- a literature which includes designs for molecular tools, ab initio quantum chemistry calculations of specific tool-surface interactions, and implementation strategies. My research colleagues and I have published many papers in this new and exciting area, and this work sharply contradicts Smalley's sweeping dismissal of the field. Smalley is just not addressing the issues. Instead, he veers off into metaphors about boys and girls in love. He describes mechanosynthesis as simply 'mushing two molecular objects together' in 'a pretend world where atoms go where you want.'"

"Actually," Merkle says, "Ab initio quantum chemistry calculations don't involve love, or mushing, or pretending. For example, a carbon-deposition reaction which a colleague and I studied using standard quantum chemistry methods moves a carbene tool along a barrier-free path to insert a reactive carbon atom into a dimer on a diamond (100) surface. The tool is then twisted 90 degrees, breaking an internal pi bond, and pulled away to break the remaining sigma bond, leaving a single carbon atom bonded to the dimer on the surface." Merkle adds, "Further computational chemistry research into fundamental mechanosynthetic reactions should be an integral component of any national nanotechnology program. Smalley's metaphors merely cloud the issues."

Samples of Merkle's publications on mechanosynthesis:

R. C. Merkle, R. A. Freitas Jr. (2003) "Theoretical analysis of a carbon-carbon dimer placement tool for diamond mechanosynthesis," Journal of Nanoscience and Nanotechnology, 3:319-324.

S.P. Walch, R. C. Merkle (1998) "Theoretical studies of reactions on diamond surfaces," Nanotechnology, 9:285-296.

R. C. Merkle (1993) "Molecular Manufacturing: Adding Positional Control to Chemical Synthesis," Chemical Design Automation News, 8,9:1.

C. Musgrave, J. Perry, R. C. Merkle, W. A. Goddard III (1991) "Theoretical studies of a hydrogen abstraction tool for nanotechnology," Nanotechnology 2:87-195.

What do enzymes have to do with MNT?

The nanomachinery of enzymes helped inspire the concept of MNT, and as discussed in the literature, enzymes and other biomolecular structures remain highly relevant to MNT research. In a lengthy technical article in the Annual Review of Biophysics and Biomolecular Structure (1994) Eric Drexler compares and contrasts "Solution Synthesis, Enzymatic Synthesis, and Mechanosynthesis", observing that mechanosynthesis (the basis of MNT) will be able to use tools providing "enzyme-like reaction environments" as a potentially useful part of a broader toolkit for controlling reactions.

In his final reply in the C&EN exchange, Prof. Smalley focuses on enzymes. He abandons his claim that mechanosynthesis must use "Smalley fingers" (which are impossible), retreating to the weaker claim that it must use wet, enzyme-like tools (which would merely be limiting). He asserts that mechanosynthesis must rely on enzyme-like mechanisms (true, for a sufficiently broad definition of "enzyme-like"), then incorrectly claims that "Any such system will need a liquid medium. For the enzymes we know about, that liquid will have to be water, and the types of things that can be synthesized with water around cannot be much broader than meat and bone of biology." These remarks -- which reveal an understanding of enzymatic chemistry that is 19 years out of date -- are apparently intended to suggest that all possible forms of mechanosynthesis must use soggy tools. They provide the only remaining grounds for Prof. Smalley's scathing rejection of current mechanosynthetic concepts. Yet Prof. Smalley's key claim -- that the enzymes we know about require liquids and water -- is simply false.

As first noted by A. Zaks and A.M. Klibanov in Science (1984, 224:1249-51), and as discussed in the 1994 Annual Review article mentioned above, even among enzymes that ordinarily work in water, many can also function in anhydrous organic solvents. Indeed, some enzymes (eg, lipase, alcohol dehydrogenase) can operate on substrates in the vapor phase, with no liquid at all. As Prof. Klibanov states in R&D Innovator (1993 2:#32), "...using an enzyme in organic solvents eliminates several obstacles that limit its usefulness in water. For example, most compounds that interest organic chemists and chemical engineers are insoluble in water, and water often promotes unwanted side reactions....Consequently, once it was established that enzymes can work in organic solvents with little or no water, R&D in the area surged."

The watery limitations on mechanosynthetic engineering suggested by Prof. Smalley are illusory -- even unmodified natural enzymes escape them. Exploring the real limits of mechanosynthesis will involve fascinating research in quantum chemistry and nanomechanical design.

What does MNT have to do with "gray goo"?

Modern MNT envisions desktop-scale manufacturing systems, not roaming, replicating "nanobots". Despite widespread misperceptions, MNT does not require the construction of autonomous self-replicating nanomachines. The Foresight Guidelines on Molecular Nanotechnology show how to steer clear of such dangers. Scale-up approaches that researchers have described in terms of "self replication" typically use devices that are merely tools under external control. This makes MNT safer than, for example, competing biological technologies.

What does MNT have to do with the dangerous self-replicating machines that Prof. Smalley fears? He seems to claim that such things are a risk solely of MNT, but this is absurd. Any sufficiently capable manufacturing technology could be used to build such devices, and the direction of nanotechnology as a whole is toward extremely capable manufacturing. Regardless of the details of future manufacturing technologies, developing dangerous self-replicating machines would require a deliberate effort with little practical motive. More realistic fears about future technologies center on the deliberate development of weapon systems, with the usual very practical motives.

False denials of the feasibility of MNT will not make genuine dangers go away, and could steer research toward a more dangerous path. Further, in a competitive world, failure to develop MNT would result in technological and economic obsolescence with results equivalent to unilateral disarmament.

Why is MNT still controversial among chemists?

MNT is a new paradigm for the future of chemistry and manufacturing, based on programmed molecular motion guided by nanomachines. Some chemists with careers tied to the old paradigm (based on random molecular motion in liquids) seem confused and threatened by this different and more powerful approach. Understanding MNT requires a bottom-up reassessment of how chemistry could work in the future. Members of the old guard instead have assured one another that MNT is "an impossible, childish fantasy" -- in short, that there is nothing to learn. Having failed to master the basic principles of MNT, they see its revolutionary promise and dangers as false, and try urgently to dismiss it. The idea nonetheless continues to gain ground because the scientific facts support it. Efforts to dismiss MNT will surely continue, however, because careers, reputations, and funding are at stake. History shows that paradigm changes take time.

What about other scientists' objections to MNT?

Other objections to MNT still circulate, but they were answered long ago. Given the extensive research accomplished in areas such as molecular manipulation, nanomechanical design, and quantum chemistry of mechanosynthesis, unsubstantiated denials of the feasibility of MNT are no longer credible. The remaining skeptics typically ignore the research literature, offer no scientific specifics, and prefer to remain anonymous. The basic research literature on MNT has withstood years of scientific scrutiny. Any major, scientifically documented error would by now be well known. The idea that there is scientific evidence against MNT is a chemists' urban legend.

What scientific breakthroughs are required for MNT?

It is often mistakenly thought that we are waiting for new scientific discoveries to validate MNT, or to make it possible. MNT, however, is an engineering concept based directly on well-established principles of chemistry, physics, and system design. The principles of MNT involve no new basic science, only new applications. Its development, like the Moon-landing program or any innovative engineering project, will require considerable trial and error learning -- that is, many routine, incremental scientific discoveries. Its development will require hard work, but not scientific breakthroughs. This is why the debate over MNT and the future of nanotechnology already has scientific answers.

Doesn't MNT sound like science fiction?

MNT promises so much that it resembles science fiction, but it matters what kind of science fiction it resembles. For decades, science fiction authors have read about scientific progress, anticipated future developments, and written about them. Examples include visions of Moon landings, nuclear bombs, and the internet. Authors have also written about apparent impossibilities -- warp drives, time travel, antigravity, and the like. But we must distinguish between ideas that are fiction because they can never happen, and those that are fiction because they have not happened yet -- between fantasy and forewarning.

Research in MNT is based on established science. It can happen, and in a competitive world, it seems hard to avoid. It resembles science fiction because it warns of enormous consequences, not because it is fantasy. To reject these forewarnings merely because they resemble science fiction would be to reject scientific evidence because it appears to have enormous consequences. This would be an enormous mistake.

Are we making progress toward MNT?

In 1959, Richard Feynman saw the achievement of his vision as "a development which I think cannot be avoided". The current world-wide surge in nanotechnology research reinforces his conclusion. Advances in chemistry, protein engineering, nanolithography, scanning probes -- all are contributing to a technology base which will enable the physical implementation of MNT systems. Though a focused effort will be required to build working systems, the distance that such an effort must cover shrinks every year. While current laboratory research in nanotechnology has no overt focus on MNT, it is well known that basic research is often useful in unintended ways. Clear articulation of research goals, however, would speed progress and improve understanding of the risks and opportunities ahead.

What can be done to move forward?

MNT promises revolutionary advances in areas ranging from medicine and materials to weapons, computers, environmental protection, and economic productivity. In a competitive world, the pursuit of MNT isn't optional. What can be done to advance understanding of the technology, and to bring the vision articulated by Richard Feynman closer to reality?

A first step will be to reject fearful, urgent, and persistent misrepresentations of the Feynman vision and its consequences. MNT is a feasible manufacturing technology of unprecedented strategic importance and practical value. False denials of the feasibility of MNT will not protect the world from its genuine risks.

Deeper understanding and practical advances will require investment in focused engineering research and development, featuring computational modeling of designs and experimental implementation of nanomechanical systems. This work will augment and build on today's broad, multi-billon-dollar program in nanoscale science and technology. It will apply current chemistry and fabrication techniques to develop the tools needed to make MNT a practical reality.

Many will benefit from an environment that accepts and encourages work centered on MNT. Researchers will be enabled to explore key technical questions. Students will find it easier to pursue advanced studies. Business enterprises will find it easier to pursue high-leverage enabling technologies. Civil society groups will find a wider audience for concerns regarding the environmental, economic, and security consequences of these technologies and where they lead.

The present irrational hostility of some scientists toward the idea of MNT obstructs both scientific progress and public discussion. In the race toward a revolutionary technology of decisive economic and military potential, the anti-research policy espoused by some leads only toward an uninformed and uncompetitive position. It is time for the scientific community to examine the facts, reject the misrepresentations, and move toward realization of the Feynman vision.