"Can we design and construct a device or scheme that can arrange atoms or molecules according to an arbitrary, user-defined blueprint? This is at the heart of the idea of the software control of matter – the creation, perhaps, of a 'matter compiler' which will interpret software instructions to output a macroscopic product in which every atom is precisely placed."
The above is not CRN's writing – it is, in fact, a product of the EPSRC, the UK's Engineering and Physical Sciences Research Council. In January of 2007, EPSRC will call together between 20 and 30 interdisciplinary researchers to brainstorm and develop research proposals toward "Software Control of Matter at the Atomic or Molecular Scale."
What's more, EPSRC has already set aside £1.5 million (almost $3 million) to fund whatever research projects come out of the two-day meeting.
The project was proposed by Philip Moriarty and will be overseen by Richard Jones. Both of them have shown a keen interest in molecular manufacturing; Moriarty has begun a collaboration with Robert Freitas, with an eventual goal of demonstrating carbon dimer deposition by scanning probe. (Freitas wants to take that all the way to direct construction of the first tiny nanofactory, and has announced a Nanofactory Collaboration.)
The EPSRC project is not directly focused solely on molecular manufacturing, but it could certainly encompass molecular manufacturing. The project description invokes both the atomic pick-and-place rapid-prototyping approach familiar to diamondoid enthusiasts, and the self-assembly approach favored by bio-philes.
If the "ideas factory" meeting achieves the broad-minded and forward-looking scope that it seems to be trying for, it has a good chance of generating at least some research that is directly relevant to molecular manufacturing. Regardless of whether self-assembly can build macroscopic products in theory, practical nanoscale fabrication techniques will probably benefit from some form of directly programmable manipulation at the nanoscale. The only way I know to make nanoscale manipulation scaleable to macroscopic levels is to use exponential manufacturing. (A possible exception is engineered bacteria, but last I heard they were still extremely hard to program.) It will be interesting to see whether the "ideas factory" considers direct nanoscale manipulation a goal worth planning for at this point. If so, and if they don't manage to find any reasonable alternative to exponential manufacturing, then they just about have to research molecular manufacturing concepts.
When I have contemplated what might kick-start a targeted molecular manufacturing project, I have imagined something very much like this "ideas factory." A few dozen diverse and creative experts, looking for ways to make direct control of atoms work; significant seed funding for whatever concepts they come up with; and all under the umbrella of a large organization (in this case, the UK government).
It remains to be seen how forward-looking the research proposals will be – the group could easily come up with £1.5 million worth of solid, cautious, worthwhile research that would not advance molecular manufacturing much. And it remains to be seen whether the group will end up focusing on the self-assembly side of the goal. But this will be an effort to watch very closely; it appears to have a chance of kick-starting major momentum toward a molecular manufacturing program.
(Hat tip to Richard Jones's Soft Machines blog, which contains a couple of details I didn't find on the EPSRC site.)