It’s clear that Richard’s efforts in establishing this Ideas Factory blog have paid off – there have been some fascinating and thought-provoking comments posted here. As Jeremy (Baumberg) states above, however, perhaps the key challenge for next week’s sandpit will lie in identifying schemes for the control of matter which can be implemented on a three to four year timescale (i.e. within the typical timeframe of an EPSRC grant). This will ultimately require the proposal of **well-defined** materials systems and manipulation protocols. Brian Wang’s and Kurt’s comments above are interesting in this regard. (However, Brian’s inclusion of the ability to generate “a complex synthetic material like a …room temperature superconductor” as one of the assessment criteria is perhaps just a little ambitious…!).
My research interests of relevance to the matter compilation theme span SPM-directed manipulation of single atoms and molecules under what might be termed extreme conditions (ultrahigh vaccum/ 4K – 300K) to self-assembly/organisation of molecules and nanoparticles deposited from solution onto solid substrates (at atmospheric pressure, room temperature). With regard to atomic/molecular manipulation using scanning probes as pioneered by Eigler et al., the bottom line – as I see it – is that SPM represents the only tool currently at our disposal which can drive and characterise fundamental mechanosynthesis reactions or postionally controlled processes.
I find much to agree with in Rob. Freitas’ post (Comment 5) in that I think that there is a lot of exciting science to pursue in the area of computer-driven positionally-controlled chemistry using scanning probes. Techniques such as inelastic tunnelling spectroscopy could be performed in parallel with dynamic atomic force microscopy/spectroscopy in order to characterise the tip structure and chemical nature during positionally-controlled fabrication of nanostructures. To date, and to the best of my knowledge, a complex 3D structure (analogous to, say, one of the quantum corrals fabricated by Mike Crommie et al. in 1993) has not yet been built with scanning probes. I’m very interested in exploring whether – **with appropriate tip functionalisation** and consideration of the (complicated) potential energy landscape and associated energy barriers -autonomous atom-by-atom fabrication of 3D nanoparticles is possible using scanning probes. Diamondoid structures and hydrogen-passivated diamond (and silicon) surfaces are likely to be of particular importance.
Hence, unlike Kurt (Comment 18 above), I believe that there is significant mileage in exploring “dry” nanotech concepts. This doesn’t mean, however, that I see scanning probe methods as being scalable to the nanofactory/molecular assembler concepts put forward by Drexler et al. (Chris and I spent a considerable amount of time discussing the “machine language” of molecular manufacturing a couple of years ago so I’m not going to repeat the arguments here.)
Bringing together my interests in both “brute force” directed assembly using scanning probes and self-assembly/self-organisation, I’m particularly interested in the “grey area” between fully deterministic positional control and directed self-assembly and self-organisation. (I realise that “directed self-assembly” is rather an oxymoronic term but bear with me…). STM tips generate intense electric fields gradients which can be used to drive the diffusion of adsorbed atoms and molecules in a given direction (Stroscio, Whitman et al. demonstrated this for Cs on GaAs(110) back in the early nineties). In liquids, many groups (including ourselves in Nottingham) are playing with dielectrophoresis as a method of tuning self-assembly of nanotubes and nanoparticles.
A grand challenge for nanotechnology is to connect the nanoscopic and macroscopic worlds. I believe that there is particular potential in combining far-from-equilbrium pattern-forming processes with molecular design and the exploitation of non-covalent interactions to tune the structure of matter across a wide range of length scales (nanometres to microns to millimetres (?!)). If self-assembly and self-organisation can be directed via external fields, pH differences etc… then there is a broad (and tunable) parameter space to explore. In terms of specific systems, I’m keen on using functionalised metal nanoparticles. Metal nanoparticles (and metal nanoparticle assemblies) also have rather interesting plasmon-mediated optical properties which, in the spirit of Richard’s “It’s all about metamaterials” post on Soft Machines, could feed into the control of photon “flow” through a structure.
While on the topic of directed- and self-assembly, Chris in Comment 4 suggests that most nanotechnologists are only comfortable with a regime whereby “Complex conditions->complex phenomena->complex output”. I’d quibble with this! It’s generally appreciated that very many systems whose dynamics can be described with rather simple differential equations and a hadful of variables (a damped, driven pendulum or Turing reaction-diffusion systems, for example) give rise to remarkably complex outputs. Hence, simple computational models (e.g. the Ising model) can give rise to complicated and rich dynamic behaviour.