Silicon Integrated Circuits

Both economic and general physics considerations indicate that the rapid improvements we have come to expect in silicon integrated circuits will saturate around the year 2010. However, fundamental physical laws indicate that it should be possible to compute with a power efficiency that is at least one billion times better than present silicon electronics.[1] The most straightforward way currently known to achieve such efficiencies are to fabricate circuits very much smaller than they are at present. Thus, there is a tremendous business incentive to invent new electronic devices and circuits that will have dimensions of the order of nanometers. In addition, new fabrication techniques will be required that can inexpensively produce and connect these devices in vast quantities. The challenges are equivalent to those faced by the inventors of both the transistor and the integrated circuit, who replaced the existing vacuum-tube and wiring technologies with solid-state switches and lithographic fabrication, respectively. In order to satisfy both requirements simultaneously, we have assembled a transdisciplinary team of chemists, physicists, engineers and computer scientists at HP Labs and the University of California Los Angeles.

Two complementary research areas relevant to future nanoelectronic systems are currently under investigation: (1) the development of quantum-state (molecular) switching devices and (2) the design of systems that will assemble themselves via kinetic or thermodynamic driving forces. Our approach for the construction of a computer based on molecular switches involves the explicit incorporation of defect tolerance, which is the capability to operate perfectly even in the presence of manufacturing mistakes in the circuit, into the design of the system.[2] This prerequisite arises from the realization that simple chemical processes cannot produce the highly complex and perfect structures that are the basis for present integrated circuits, but will instead give rise to highly ordered but defective systems. However, by constructing the appropriate molecular circuitry, arbitrary complexity can be programmed into a highly regular structure and at the same time any defects can be avoided by setting the states of appropriate switches.

Our research group and our partners at the University of California Los Angeles have recently demonstrated that it is possible to construct molecular switches in a solid state device that can be set and read electronically.