This paper is available on the web in Microsoft Word for Windows 2.0 format at ftp://quake.unr.edu/pub/gillett/extract.rev.
I have previously suggested that the molecular assembly of nanostructured materials promises to be a major driver toward achieving full-blown molecular nanotechnology, by providing the short-term financial incentives for the enormous scale of research and development required. Perhaps the most compelling set of applications stems from a single, fundamental technological problem: separating out a few of one type of atom or molecule from a huge (and arbitrary) background of other types. Viewed in one way, this is the fundamental problem of pollution control, as well as of desalination; viewed in another way, it is the problem of resource extraction.
Traditional element separation, however, employs the thermal partitioning of elements between coexisting phases. Although simple, this is extremely expensive: high energy costs result from the prodigious application and extraction of heat to force phase changes, and extraction and preparation of the reagents required in many processes further add to costs. Thermal partitioning is intrinsically dirty, too, because partitioning is never complete, and the combustion used in most cases to furnish heat adds yet another source of pollution. Finally, for resource extraction thermal partitioning requires ores, naturally occurring feedstocks in which the concentration of the desired element is already anomalous. The extraction of desired elements from very low-grade sources such as common rock remains impractical.
Because it already involves dealing with low concentrations, pollution control largely relies on non-thermal approaches to element separation, and in many cases has driven what has become a large and growing market. Several such approaches exist: semipermeable membranes (for electrodialysis or reverse osmosis), ion exchange resins, molecular sieves (zeolites and zeolithoids), ion pumps (as in biological systems), and so on. As all such approaches fundamentally move individual atoms differentially, the materials used in them are fundamentally nanostructured. Because they are isothermal, these approaches can be vastly cheaper than thermal extraction. Indeed, natural biological systems show what is possible: vertebrate kidneys, which isothermally select and extract a few electrolytes from a background of many other electrolytes, are perhaps the most outstanding example.
Current fabrication of the materials used in these element-separation techniques leaves much to be desired, however, as it relies on conventional wet-chemical syntheses consisting largely of ad hoc "recipes". The result is extremely limited control over the final product. Many undesired byproducts typically also result, and finally, more sophisticated (and complex) structures simply cannot be fabricated by such traditional "shake and bake" methods. The resulting high costs, moreover, mean that non-thermal techniques remain largely impractical for resource extraction; even seawater desalination is generally not cost-effective. By contrast, relatively unsophisticated, repetitive molecular assemblers--"molecular looms"-- have the potential of fabricating such nanostructures routinely, cheaply, and with much more sophisticated and reproducible patterns.
Many billion dollars' worth of applications lie in pollution control alone, and the increasingly stringent restrictions on wastewater content indicate this market will only increase. Thus, perhaps the "breakthrough" application of nanotechnology will be cheap, molecularly designed and fabricated "industrial kidneys." This would be somewhat ironic, as the original Limits to Growth scenarios of the early 1970s assumed that escalating pollution ultimately rendered any technological advances counterproductive--in retrospect the worst assumption of those models!