Foresight Institute Logo
Image of nano

How to save $149,850 per nanotechnology experiment

Atomistix has just expanded to the U.S. at the Innovation Center Denmark in Palo Alto, not far from Foresight. I missed their open house on September 28, but I see that their CEO Thomas Magnussen explained the value of computational modeling in nanotechnology:

Over time, the price of nanotechnology experimentation has gone up while the cost of modeling has dropped dramatically. Based on calculations from Dow Chemical, Magnussen now estimates that a typical experiment costs 150,000 USD versus 150 USD for the same experiment conducted in virtual reality.

Sounds good to me. Perhaps one of our readers can explain the differences between the Atomistix product and the Nanorex product, often mentioned in this blog.

Meanwhile, they’re saving a ton of money over at Georgia Tech: Also on September 28 was another event in nanotech computer modeing, the publication in the Journal of Physical Chemistry — a top chemistry journal — of the paper “High-level Ab Initio Studies of Hydrogen Abstraction from Prototype Hydrocarbon Systems” by Berhane Temelso, C. David Sherrill, Ralph C. Merkle, and Robert A. Freitas Jr.

For his work on this paper, Temelso won this year’s Foresight Distinguished Student Award. A team to watch on the pathway to mechanosynthesis. —Christine

4 Responses to “How to save $149,850 per nanotechnology experiment”

  1. NanoEnthusiast Says:

    It is my understanding that the Nanorex software has specific tools relevant to MNT work, i.e. tools to help define MNT parts (like bearings) parametrically that can then be optimized at a later stage to reduce chemical instabilities. From what little I can tell, this package would not have any of that. This package however, has tools relevant to general crystalline growth and nanotube creation as basic primitives.

    I hope that there is, or will emerge, a standard file format for this kind of work. It does no one any good to have a tower of Babel situation where different research communities can’t use each others files. Also it is important (in order to head off any skepticism) to make sure that any interesting discoveries made in NanoEngineer-1 can be easily duplicated in other software applications.

  2. John Novak Says:

    I hope that there is, or will emerge, a standard file format for this kind of work. It does no one any good to have a tower of Babel situation where different research communities can’t use each others files.

    I respectfully disagree.

    At this stage of the game, the best format for nanotech simulations is not known. The virtue in the babel situation is this: by not agreeing to a particular set of files and formats and methodologies, the babel promotes experimentation and exploration of the design space, making it more likely that over time, even better approaches than the ones we have now will be discovered. Early centralization and pre-mature convergence are bad things, not good things.

    Moreover, I find it exceedingly unlikely that there will be found a universally “best format” for these sorts of things, any more than there is a universally “best programming language” even after decades of research into computer science. The lesson taken from computer science, rather, is that different goals are enabled by different techniques.

    We don’t want total anarchy, no, but we don’t want total conformity, either.

  3. Carlos Bray Says:

    I think it’s a moot point because once there was a need for translation from one format to another some guy will make a program for just that very thing. The exact same way there are programs that allow you to change the format of a image file.

  4. NanoEnthusiast Says:

    Perhaps we need at least the option to export in a universal format, but still have formats unique to each software package. Most word processors and 3d animation packages have data formats that are used only by the vendor, and yet you can still export to something neutral and continue where you left-off in another package. At any rate, NanoEngineer-1 is supposed to be released as open source, so there will be plenty of documentation regarding their file format(s) in the source code. Licensing issues permitting, there will be ways for other software to import and export NanoEngineer-1 files either natively or through plug-ins.

Leave a Reply