2005 APS March Meeting
Monday–Friday, March 21–25, 2005;
Los Angeles, CA
Session L5: Emerging Devices and Materials for the Microelectronics Industry
2:30 PM–5:30 PM,
Tuesday, March 22, 2005
LACC
Room: 502B
Sponsoring
Unit:
FIAP
Chair: Alex Demkov, University of Texas, Austin
Abstract ID: BAPS.2005.MAR.L5.1
Abstract: L5.00001 : The Search for New Information Processing technologies*
2:30 PM–3:06 PM
Preview Abstract
Abstract
Author:
Ralph K. Cavin
(Semiconductor Research Corporation)
Our society has benefited from the ‘Golden Age of Electronics’
for the last half century. The ubiquitous transistor, in its
many manifestations, has enabled an explosion of capabilities
in information processing, communications, and sensing that has
spurred exponential growth in performance-benefit ratios.
Much of the credit for this progress is due to the continued
scaling of the silicon integrated circuit (IC) components and
to the associated efficient fabrication processes that have
made the IC affordable. There is a growing realization, from
simple physics arguments, that as minimum features sizes
approach the ten nanometer regime, scaling will very likely
slow and eventually end. This doesn’t mean that the MOSFET
will disappear, but more likely that it will need to be
supplemented by other device and interconnect technologies if
the exponential gains are to continue.
In this talk we discuss the basis for the projected limitation
of scaling of charge-based devices for logic and memory
devices. We argue that a fundamental consideration for all
devices, including those based on charge, relates to the
capacity to manage heat generated by circuit operation. Our
preference is for devices that operate at room temperature
since the energy costs for cooling the devices must also be
charged against the overall system energy consumption.
(Cooling costs increase as a power of the difference between
the ambient and the target temperature.) Therefore we seek new
state variables to serve as an alternative to electrical charge
for future information processing technologies. These
technologies must provide the potential for sustaining
exponential performance-cost benefits with time. The search
must not only focus on device structures but on the underlying
materials and process technologies that enable these
structures. Indeed, to obtain extremely scaled CMOS, new
materials and processes must also be developed. In this talk,
we survey some of the candidates for replacements/supplements
for/to the MOSFET and give a status report on the status of the
search.
We also briefly discuss the problem of design in the far
nanometer regime where device variability is likely to be
significant. What design constraints must be employed to
ensure that manufacturing yields are high, given broad
tolerance margins for the device characteristics? Variability
is a growing problem in extremely scaled CMOS what is learned
from these applications will likely benefit replacement
technologies as well.
*In Collaboration with Victor Zhirnov
To cite this abstract, use the following reference: http://meetings.aps.org/link/BAPS.2005.MAR.L5.1