A while back we mentioned an article in Nanotechnology Law & Business in which the author seemed not to have done his homework. A rebuttal letter should be in the September issue, and we have reproduced it below. Sure enough, the NLB author did not do his homework at all, apparently. Tisk, tisk! Where were the reviewers on this one? —Christine
In “Nanoassemblers: A Likely Threat?” Martin Moskovits claims to “focus almost exclusively on the prospects of nanoassemblers” while omitting even basic references in the area.
This letter will focus on just a few of the fundamental errors made by Moskovits, rather than attempting to critique the numerous flaws and omissions in his paper.
Moskovits spends a few pages discussing Maxwell’s Demons (which violate the second law thermodynamics) and mentioning perpetual motion — implying that “nanoassemblers” also violate the second law but without stating how this occurs. He then advances three “reasons” why “nanoassemblers” and “Maxwell’s Demons” are both “unlikely.”
1) “…the nanobot’s limbs will jitter and their accuracy as an assembler will be faulty” because of thermal noise and quantum uncertainty. Both issues have been (repeatedly) addressed in the literature. Standard literature citations (not mentioned by Moskovits) include chapters 3 and 5 of Nanosystems (Drexler, 1992). For a discussion available on the web, see Nanomedicine Vol I (Freitas, 1999) http://www.nanomedicine.com/NMI/2.1.htm. For more specific design proposals for molecular positional devices along with an analysis of thermal noise and positional uncertainty, see chapter 13.4 of Nanosystems or “A New Family of Six Degree Of Freedom Positional Devices” (Merkle, 1997 http://www.zyvex.com/nanotech/6dof.html ).
Also relevant is “Theoretical Analysis of Diamond Mechanosynthesis. Part III. Positional C2 Deposition on Diamond C(110) Surface Using Si/Ge/Sn-Based Dimer Placement Tools” by J. Peng et al. available on line at http://www.molecularassembler.com/Papers/JCTNPengFeb06.pdf. This paper specifically discusses the molecular dynamics simulation of positional uncertainty in the mechanosynthetic context, providing specific tolerances in each Cartesian and rotational direction of the tooltip that take full account of zero point and thermal fluctuations.
For a simpler and more compact discussion for a general audience, see “That’s impossible! How good scientists reach bad conclusions” at http://www.zyvex.com/nanotech/impossible.html in the sections labeled “Thermal Noise” and “Quantum Mechanics.”
2) “For a nanobot to pick up a desired molecule from the environment it will need to have on-board analysis capability.” Moskovits goes on to argue that quantum uncertainty will prevent binding to a desired molecule. The proposals in the literature for selective binding to relevant feedstock molecules again go back to Nanosystems (chapter 13). This is also discussed in Nanomedicine Vol I Chapter 3 (http://www.nanomedicine.com/NMI/3.5.htm) as well as “Binding sites for use in a simple assembler” (Merkle, 1997, http://www.zyvex.com/nanotech/bindingSites.html). It should also be noted that biological systems have been binding to and utilizing various molecular feedstocks for some billions of years.
3) Moskovits third argument claims that molecules “released” during synthesis will float away. Again, Nanosystems discusses mechanosynthetic reactions [chapter 8] making it clear that proposed reactions do not have such imagined failure modes. Papers on various reactions useful for mechanosynthesis have appeared in the literature (see http://www.molecularassembler.com/Nanofactory/AnnBibDMS.htm for a bibliography).
Moskovits also argues that mechanosynthesis is “inefficient”. He questions the ability to build macroscopic structures in a reasonable period of time from molecular building blocks. We note, however, that ribosomes exist and are evidently able to build macroscopic structures from molecular building blocks (amino acids), thus they must be “efficient.” A variety of “efficient” manufacturing architectures are discussed in “Kinematic Self-Replicating Machines” (Freitas and Merkle, 2004, http://www.molecularassembler.com/KSRM.htm ). Of particular interest is convergent assembly, also discussed in chapter 14 of Nanosystems as well as “Convergent Assembly,” (Merkle, 1997, http://www.zyvex.com/nanotech/convergent.html).
We look forward to hearing Moskovits’ response after he has had time to read a few of the relevant references.
Ralph C. Merkle, Ph.D.
Senior Research Fellow
Institute for Molecular Manufacturing