Computer Systems Laboratory Colloquium

4:15PM, Wednesday, October 9, 2002
NEC Auditorium, Gates Computer Science Building B03
http://ee380.stanford.edu

Molecular electronics:
Defect Tolerance, Chemical Fabrication and Quantum-State Switching

R. Stanley Williams
HPLaboratories, 1501 Page Mill Rd., MS 1123, Palo Alto, CA 94304 (USA
About the talk:

Both economic and physics considerations indicate that the rapid improvements we have come to expect in silicon integrated circuit technology could saturate around the year 2010. However, fundamental physical laws reveal that it should be possible to compute with an electrical power efficiency that is at least one billion times better than present transistor electronics. The most straightforward ways currently known to achieve such power efficiencies are to fabricate much smaller circuits and to minimize the number of transistors they contain. Thus, there is a tremendous business incentive and scientific challenge to invent new electronic devices that will have critical dimensions of the order of nanometers and new fabrication techniques that can inexpensively produce and connect these devices in vast quantities. These problems are equivalent to those faced last century by the inventors of both the transistor and the integrated circuit, who replaced vacuum-tubes and wiring technologies with solid-state switches and lithographic fabrication, respectively. In order to transform molecular electronics into a true technology in the next decade, we have assembled a trans-disciplinary team of chemists, physicists, engineers and computer scientists at Hewlett- Packard Labs to invent quantum-state switches and chemical assembly techniques.

Three complementary research fields must be integrated to achieve this goal:

  1. the design of defect-tolerant circuit architectures,
  2. the creation of molecular switching devices and
  3. the development of inexpensive chemical fabrication techniques.
Our approach to molecular electronics explicitly recognizes that it will be prohibitively expensive to build a complex and perfect structure with nanoscale dimensions because of statistical mechanical limitations. Thus defect tolerance, which is the capability to operate perfectly even in the presence of manufacturing mistakes in a circuit, is essential to any scheme for nanoelectronics. Simple chemical processes cannot produce the highly complex and perfect structures that are the basis for present integrated circuits, but will instead give rise instead to ordered but slightly defective systems. However, by constructing crossbar arrays with molecules at the junctions that act as nonvolatile electronic switches, arbitrary complexity can be programmed into a matrix structure and at the same time any physical defects can be avoided by setting the on/off states of appropriate switches.

Our research group at HPL has recently demonstrated molecular switches in a solid-state device with 10,000 to 1 on-to-off resistance ratios that can be reversibly set and read electronically. These switches can be used as the basis for storing bits with values of 1 or 0 in a memory, or they can be configured to create circuits that perform simple logic functions. We have built and tested memory circuits that represent the world's highest density electronically addressable memory. In the process, we have discovered new physics, in that electron transport through organic monolayers is qualitatively different from that through inorganic thin films, and we speculate that these differences are the cause of the unexpectedly large electronic switching effects we observe. This program clearly demonstrates how fundamental research in a corporate laboratory can be a strategic asset for the parent company, and how it is possible to mix curiosity-driven discovery in the physical sciences with invention in the electronics industry by the proper choice of research focus.

About the speaker:

R. Stanley Williams is HP Fellow at Hewlett-Packard Laboratories and founding Director (since 1995) of the Quantum Science Research (QSR) group. The QSR was established to prepare HP for the major challenges and opportunities ahead in electronic device technology as features continue to shrink to the nanometer size scale, where quantum mechanics becomes important. He received a B.A. degree in Chemical Physics in 1974 from Rice University and his Ph.D. in Physical Chemistry from U. C. Berkeley in 1978. He was a Member of Technical Staff at AT&T Bell Labs from 1978-80 and a faculty member (Assistant, Associate and Full Professor) of the Chemistry Department at UCLA from 1980 - 1995. His primary scientific research during the past twenty years has been in the areas of solid-state chemistry and physics, and their applications to technology. This has evolved into the areas of nanostructures, chemically-assembled materials, and molecular electronics. His awards for scientific and academic achievement include the 2000 Julius Springer Award for Applied Physics and the 2000 Feynman Prize in Nanotechnology. He has been awarded seven US patents, has published 201 papers in reviewed scientific journals, and has written general articles for technical and business publications.

See the following links for further information:

http://www.hpl.hp.com/research/qsr/index.html

http://www.hpl.hp.com/features/stan_williams_interview.html

http://www.forbes.com/tool/html/99/jun/0614/feat.htm

Contact information:

R. Stanley Williams
HP Laboratories
1501 Page Mill Rd., MS 1123
Palo Alto, CA 94304 (USA)