• Google
    This Blog Web

October 2011

Sun Mon Tue Wed Thu Fri Sat
            1
2 3 4 5 6 7 8
9 10 11 12 13 14 15
16 17 18 19 20 21 22
23 24 25 26 27 28 29
30 31          

RSS Feed

Bookmark and Share

Email Feed



  • Powered by FeedBlitz

« Nanotech and the Pentagon | Main | Enabling Technologies »

April 23, 2004

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

Kurt

For the past two weeks, I have been conducting a direct market campaign for my instruments (an optical thin-film tester and a surface acoustic wave system). Going through all of the university websites, I noticed that almost every university that has a chemistry and materials science program are working on self-assembly and bio-memetic (wet) nanotechnology.

I find it very difficult to believe that we will not have a comprehensive "wet" nanotechnology capability within 20 years. The problem I have with "dry" nanotech is that the currently proposed schemes to not take into acount the use of brownian motion. Browning motion is inherent to all molecular level systems and is the reason why all of the proximinal probe manipulations to date have all be done at cryogenic temperatures, not very useful for real-world systems.

If someone comes up with a "dry" nanotech that incorporates brownian motion as part of the operating principle, I would be much more convinced of its feasibility than I am now.

Prof. Smalley does alot of hand-waving in his arguments against the possibility of "dry" nanotech. However, if you read through the hand-having, the core of his argument is the issue of brownian motion.

There is a professor Seeman at NYU who has come up with a list of 10 technical hurtles that msu be overcome for a feasible nanotech to be developed. Does anyone here have a list of these hurtles?

Richard Jones

Chris, a tedious technical comment. Your statement that "Crystals are existence proofs of flawless arrangements of atoms" isn't quite right; at finite temperatures the second law says you have a finite number of point defects - vacancies and interstitials - with the fraction of sites with defects scaling like exp(-ev/kT) where ev is effectively the energy of cohesion of the solid per atom. These point defects don't disrupt the perfection of the long-ranged order (at least in 3 dimensions - the story is different in 2). One might think they aren't relevant in MNT because the typical system sizes are small enough for the number of equilibrium defects to be very small. But if you start having strained structures the exponential dependence on binding energy might start to make defects a problem.

michael vassar

Actually Chris, you did hear the thermalization argument before, in a slightly different form, when you used it correctly against a proposal that I made. I then salvaged the proposal by proposing a method for phonon scattering and energy dispersal, but my later research showed that enabling technologies were still needed. The details are different in this case though, and thermalization doesn't present a problem.

Professor Seeman's 1st name is Nadrian
His e-mail is ned.seeman at nyu.edu but his web-page doesn't work.

Chris Phoenix, CRN


To Kurt, re Brownian motion:

I don't know where the rumor started that Drexler's designs don't use Brownian motion. They do.

Brownian motion is useful. For example, it can provide diffusive transport. It can retry reversible molecular matching operations. And it (or more precisely, heat) can, if I'm not mistaken, lower friction--not just by reducing viscosity but by providing a denser collection of accessible states.

Retrying molecular conformations is mentioned in sorting rotor binding (Nanosystems 13.1). Whole mechanical systems moving beneficially under thermal noise is invoked to allow a "stuck" sorting rotor to reverse itself by rotating backward under thermal noise (Nanosystems 13.2.1.d).

Thermal noise is also invoked in Nanosystems in the chapter on friction: 10.3.5 "Static friction" begins with the statement, "Where delta-curly-V-sub-barrier << kT, static friction is effectively zero." T, or temperature, is a parameter of thermal noise.

Diffusive transport appears to be less efficient than some alternatives; compare for example Nanomedicine 3.2.1 vs. 3.4.3. But it could certainly be used if anyone wanted to.

Chris

Chris Phoenix, CRN


To Richard, re entropy:

This is truly an exception that proves a rule. The back of my envelope says that C-C bond energy is 154 kJ/mol, or 256 zJ/bond, or 512 zJ/atom. That over kT300 is 124. e^-124 is 10^-54 defects per atom. So at room temperature, there'd be about one entropy-required defect in every 1.6x10^28 kg of diamond.

Chris

Chris Phoenix, CRN


Prof. Seeman has given me permission to post his "Top 10" list. Here is what he sent me:

"""""
These are the challenges for structural DNA nanotech, as I see them.

Those with two stars have seen major progress, those with one have seen a beginning.

None are completely resolved.

[1] TO EXTEND 2-D RESULTS TO 3-D WITH HIGH ORDER -- Crystallography; Nanoelectronics.

[2] TO INCORPORATE DNA DEVICES IN 2-D AND 3-D ARRAYS -- Nanorobotics. *

[3] TO INCORPORATE HETEROLOGOUS GUESTS IN LATTICES -- Nanoelectronics; Crystallography. **

[4] TO EXTEND ALGORITHMIC ASSEMBLY TO HIGHER DIMENSIONS -- Smart Materials; Computation. **

[5] TO ACHIEVE ASSEMBLIES WITH HIERARCHICAL CHARACTER -- Complex Materials. *

[6] TO ACHIEVE FUNCTIONAL SYSTEMS -- Active Materials; Sensor Systems. *

[7] TO INTERFACE WITH TOP-DOWN METHODS AND THE MACROSCOPIC WORLD -- Nanoelectronic Reality.

[8] TO INCORPORATE COMBINATORIAL APPROACHES IN THE DESIGN -- Diversity; Programmability. *

[9] TO PRODUCE SYSTEMS CAPABLE OF SELF-REPLICATION -- Economy; Evolvability. **

[10] TO ADVANCE FROM BIOKLEPTIC SYSTEMS TO BIOMIMETIC SYSTEMS -- Chemical Control.
"""""

Janessa Ravenwood

This is very useful, Chris. Maybe you could supply us with an annual "Project Progress Chart" for developing MNT with different graph lines for each of the necessary steps/phases/components necessary to pull the whole thing off. How about it?

Chris Phoenix, CRN

Just re-read this discussion. Quick comments:

Richard, you said strained structures might have a higher entropy-induced error rate, and I didn't really answer that--I just agreed with you that it wouldn't affect diamond. But a 10^-15 error rate per atom corresponds to a bond energy of just 140 zJ/atom. That's really small, and you could easily avoid building something that strained. (Drexler covered the failure rate of strained bonds in Nanosystems.) It's easy, and common, and pointless, to find examples of things that won't work. Unless those examples imply that nothing in the space will work--and I've never seen that happen in molecular manufacturing.

Janessa, I don't think we know enough yet to go into as much detail for (diamondoid) MNT as Seeman did for DNA. The best I can do is guesstimate, hopefully within an order of magnitude, how much it'll cost to develop it in the next N years. And of course the graph lines will be next to useless in forecasting, since their future progress will depend on both funding and new insights.

Chris

The comments to this entry are closed.