7.3 Molecular Electronic Devices

The attempts to realize a molecular electronic device arose from the needs to develop a next-generation electronic device. Conventional electronic devices are based on semiconductor technology. Very narrow lines are drawn on the silicon wafer that function as electric circuits. The magnitude of integration depends on how narrow the lines are. Present technology using photolithography allows lines narrower than a submicrometer, but with this size, the flow of electrons cannot be expected to work correctly. As long as we rely on a silicon-photolithography "style", growth of integration will reach a limit.

In 1974, Aviram and Colleagues (Aviram and Ratner 1974) proposed the idea of a molecular electronic device, as did Carter in 1982 (Carter 1982). Their idea was to synthesize molecules that show electronic functions. Electro-rich molecules are ready to give electrons to other molecules when conditions allow, and these molecules will play the role of electron donor. Some molecules are poor in electrons and tend to accept electrons from their environment. When the two kinds of molecules are separated with an insulating molecule, forming a donor-insulator-acceptor structure, this could theoretically function as a rectifier when sandwiched by two electrodes. This idea was not, unfortunately, followed by experimental proofs until now. The main barrier to this concept has been the difficulty of measuring a signal from only one molecule. The key to opening the gate is to mimic biomolecules.

As mentioned earlier, proteins recognize their counterpart specifically inside the cell, and they make a very narrow but safe passway for electrons in a molecular size (several tens of nanometers). If we succeed in inventing a good method to come in contact with each molecule, and also a good method to assemble them as we want, the development of a biomolecular electronic device (biochip) will proceed. With a scanning tunneling microscope, the electrical characteristics of individual atoms can be measured. The transposition of individual atoms to desired locations is also possible. Theoretically, a device based on the electrical characteristic of only one molecule could be possible, but the problem is that such a device would be extremely large. A current topic of discussion is how to miniaturize STM technology down to the point that it could be used with high consistency.

Some biomolecules work as a rectifier. For example, an accumulated LangmuirBlodgett membrane of flavin-porphiline prepared on an electrode showed the behavior of a rectifier. Ueyama, Isoda, and Maeda (1989) used six layers of LB membrane of ruthenium complex prepared on a gold electrode; two layers of organic compounds were formed over it. With this system, the electron transfer rates between each layer and electrodes were measured. The rate of one direction was much larger than the opposite way. With this result, the system acted as a rectifier.

Aizawa and coworkers (Aizawa et al, 1989) reported a system using a conductive polymer-enzyme conjugation that showed a molecular switch property. Enzyme activity was controlled by the potential of an electrode on which the conjugation was prepared.

Our technology level is now sufficient to create several submillimeter biosensors. This will be a key for the realization of future biocomputers. The combination of the work on microbiosensors and that of molecular electronic devices will be the only way to reach the goal.