STMFMs and Molecular Computing

The ability to use multi-tip STM/FMs for mechanical chemistry and as electrical interfaces might solve two formidable problems in the development of molecular computing devices as envisioned by Carter (1983): 1) the synthesis of prototype devices and 2) making individual, reliable, electrical connections to them for testing.

Arrays of nanoscale electrodes or STM tips comprising sub-micron scale assemblies may be extremely useful if their position can be precisely and rapidly regulated. Schneiker (1987) has suggested that optimal utilization of piezo-ceramic materials may be used for parallel scanning of nanoscale arrays. The piezo-ceramic "biomorph-bender" described by Dieter Pohl and colleagues (1986) 'may be used to manipulate such arrays for fast parallel reading and writing of nanoscale array chips for mass storage and processing of information (Figures 10.14 and 10.15, Schneiker, 1987).

STM/FMs could be used for more conventional ultraprecise circuit or component trimming and repair operations. The extremely accurate positioning systems of STM/FM could be exploited for minimal scale wirebonding systems. Other STM derived applications could include inspection of, and electrical and mechanical interfacing to, superlattice quantum well devices or "quantum dot" devices on the scale of cubic nanomenters. In 1959 Feynman noted that using his strategy it should eventually be possible to make what are now called superlattices and that they would probably have some very interesting properties (Feynman, 1961). Even though STM/FMs may not ultimately be used to mass produce such devices, they could be used to construct prototypes, help characterize test devices, and be used to optimize them. STM tip induced sputtering (Binnig, Gerber, Rohrer and Weibel, 1985) might be used to etch or mill ultrafine conductors for such devices. The availability of even small numbers of such extremely high performance devices would have great instrumentation value as interfaces and transducers (Figures 10.16 thru 10.19).

Computing components each consisting of a few atoms might use quantized energy levels or spin effects (Feynman, 1960); if such devices could be designed and assembled, they could be thousands of times faster than conventional devices used in today's computers. Other aspects of quantum computers are discussed in Maddox (1985), and other interesting references may be found in Schneiker (1986). Efficiently interfacing these and other such devices to the outside world in a manner that can effectively utilize their potential computing bandwidth presents some interesting problems in clocking, input/output, and other areas. Schneiker claims that the optical electronics technology suggested by Javan (1985, 1986) might be useful in these and other nanoapplications as well. Using micro and nano-antennas coupled to laser radiation, intense, localized high frequency electric fields modulated by the laser beam's intensity, phase, and polarization could control and power arrays of STM/FM constructed nanoscale devices. This same idea could be used to control, power, and communicate with swarms of mobile nanorobots.

Figure 10.14: Computer mass storage for fast parallel reading and writing "nano-film" coated chips. Ultimately, electrodes could be replaced with micro-STMs (Figure 10.15) for accessing individual molecular devices. A simple system might also be configured for highly parallel nanolithography. By Conrad Schneiker (Schneiker, 1986).

Figure 10.14: Computer mass storage for fast parallel reading and writing "nano-film" coated chips. Ultimately, electrodes could be replaced with micro-STMs (Figure 10.15) for accessing individual molecular devices. A simple system might also be configured for highly parallel nanolithography. By Conrad Schneiker (Schneiker, 1986).

Figure 10.15: A micro-STM formed on a silicon substrate; thousands of these structures may be placed on a single silicon chip. By Conrad Schneiker (Schneiker, 1986).
Figure 10.16: Three STM tips in configuration for tunnel modulation experiment. By Paul Jablonka (Schneiker and Hameroff, 1987).
Figure 10.17: Electrochemical tunnel switch/atomic memory device. By Conrad Schneiker (Schneiker, 1986).

During recent talks on his quantum computing ideas, Feynman briefly speculated on a simple possibility for making nanocomputer components: use STM tips to make tiny holes in very thin metal sheets, thus forming grids for tunneling "nanovacuum tubes," perhaps around 3 to 10 nm in size, or smaller (Feynman, 1985, 1986). Analogous, but much larger (down to 100 nm scale) devices with calculated picosecond range switching speeds have been proposed by Shoulders (1965). Although he considered even smaller and faster devices, limitations of electron beam micromachining technology at that time prevented further size reduction (Shoulders, 1986). STM/FMs could solve that problem and many others. Indeed, the Naval Research Laboratory is now studying subpicosecond (thousandth nanosecond) submicron vacuum tubes (Robinson, 1986).

Figure 10.18: STM-induced/detected molecular conformation change. By Paul Jablonka (Schneiker and Hameroff, 1987).
0 0

Post a comment