Design for a nanorobotic positing device, by K. Eric Drexler, Nanosystms, Copyright © 1992. This material is used by permission of John Wiley & Sons, Inc.

Dateline: March 9, 2001

The concept behind molecular nanotechnology is to fabricate devices where discrete atoms and molecules are the components of molecular machinery that can be controlled by external means [1]. That such a concept is realisitic can be seen in biological systems, in which molecular components such as nucleic acids, proteins and membrane components work together to carry out seemingly miraculous functions such as photosynthesis [2], thought [3] and motor control [4]. Despite the obvious existance of a biomolecular nanotechnology [5], the enabling device for nanotechnology termed "the assembler" [6] has yet to be shown feasible. Many people share a vision of a nanotechnology where a nanorobotic "assembler" (see figure) strategically picks up and positions atoms or molecules in a specific place. However, researchers who have mastered the art of moving atoms [7] and molecules [8] agree that such an approach to molecular manufacturing is likely inferior to other fabrication methods such as self-assembly [9] and templating [10], the very same methods used by biological systems to assemble biomolecular components. Thus it seems that a more realistic assembler would make optimal use these methods.

I have redefined the term "assembler" as: "A chemical device that given certain atomic or molecular inputs (starting materials) can output a specific molecular structure or aggregation." Plenty of methods are available for organizing atoms and molecules. Biomineralization is a method used by living organisms to form crystaline structures such as bones and shells, and researchers are learning how to do it in vitro [11]. Templating on the nanoscale can range from Immuno-PCR [12] to molecular imprinting [13], all molecularly precise yet sufficiently large scale enough for us to direcly interact with and control in vitro. A plethora of in vitro self-assembly techniques are also well established, including but not limited to: thiol assembly [14], Langmuir-Blodgett films [15], and even many types of biomolecular self-assembly [16]. Like templating, self-assembly is not limited to the nanoscale, and can theoretically be used to assemble components of any size between the macro- and nanoscales simply by strategically mixing various components [17].

An assembler is now a realistic possibility because single molecule detection methods [18] give us the ability to directly observe the assembly process and thus to adjust the macroscale parameters that affect it, e.g. temperature, concentration and timing. One approach towards the development of an assembler would be to integrate microfluidic devices with single molecule detectors in order to form a lab-on-a-chip [19]. Such a device would be far faster and efficient than a macroscale lab since feedback from the single molecule (or single assembly) detectors could be used to control the assembly parameters. Such an assembler would be specific only to certain types of assembly, and thus would not be a "universal assembler," allowing us to put any atom wherever we want. However, self-assembly and templating allow far more large scale molecular precision than any other means and pursuit of this redefined assembler is a promising approach towards the development of a safe and realistic molecular nanotechnology.

Acknowledgements: Thank you Rainer Ellerbrake for asking the question, "Are you a self-assembler?"