|
|
| Silicon
Nanodefects -- Could
the Future of Nanoelectronic Manufacturing Processes be
Facing an Old
Fundamental Silicon Problem? |
|
Ed Wijaranakula, Ph.D.
Chief Technical Officer, Infotix
Systems, Inc. - June 15, 2004
Page
1 of 1
|
|
Last March,
the
European Commission announced the first investment of
$30 million on a 45-nm NanoCMOS project with the goal to
become the world's leader in nanoelectronics in the next
two years. The project initiative involves various
European-based universities,
|
research
labs and companies including Infineon (NYSE ADR:IFX),
Philips and STMicroelectronics (NYSE:STM), Europe's
largest chipmaker. The U.S. chipmaker, Texas Instruments
(NYSE:TXN) announced this month that the company will
expand its relationship with the Belgium-based IMEC
(Interuniversity MicroElectronics Center) research
center by becoming a core member of their sub-45nm CMOS
research program.
Although an innovation in
late 2001 by research scientists at Lucent
Technologies' Bell Laboratories (NYSE:LU) showing the capability of
building nanotransistors from a
|
|
| a single molecule of organic (carbon-based) material known as
"thiols" [1] has demonstrated U.S. leadership
in nanoelectronic technology, Mr. John E. Kelly III, senior VP
and group executive of the IBM Technology Group (NYSE:IBM), warned
early this month at the
Semiconductor Industry Association (SIA) meeting
that the semiconductor engine driving productivity gains
would slow down, or possibly stall if the U.S.
semiconductor industry doesn't act now.
Organic
materials such as "thiols" or
semiconducting carbon nanotubes [2] could one
day be used in nanofabrication of
microprocessors and memory chips, but at the
present time, silicon appears to be the only
commercially feasible material-of-choice. Fundamentally, silicon used in today's
microelectronic device fabrication contains
various types of intrinsic and extrinsic
defects. The intrinsic defects, silicon
interstitials and vacancies, are formed during
silicon crystal solidification, while extrinsic
defects are either intentionally or unintentionally
added into silicon during the manufacturing
process. Well-known extrinsic defects are
oxygen, boron, phosphorus as well as transition
metals including Cu, Ni and Fe.
Intrinsic defects usually aggregate during
silicon solidification into nanodefects, having
a size as small as 50-nm. These nanodefects are
known to degrade the integrity and cause
structural damage in sub-micron devices.
Isolated intrinsic defects with charged states
as well as transition metals produce
transistor noise as the defects are capable of
trapping electrons and holes in the transistor
active region. Problems associated with silicon
nanodefects and isolated charged defects could
become worse for the nano-device.
Although the formation of silicon
nanodefects can be suppressed by incorporating a small amount
of nitrogen (1E15 atoms/ cm-3 range) into silicon, research
scientists at Toshiba Ceramics found that nitrogen
generates several new classes of silicon nanodefects which are not
well characterized [3]. Semiconductor companies like Intel Corporation (NASDAQ:INTC)
and Matsushita (NYSE ADR:MC) [4] have developed
high-speed and low-power nanochip technology by introducing
a SiGe strained layer, ~ 15 nm thick, into the N-channel of
the MOS transistor. Compressive stresses induced by
germanium (Ge) atoms in the SiGe nano-layer lowers the
energy bandgap, which in turn increases the electron
mobility across the channel. As
reported by Intel, the strained silicon process has a high
level of defects [5]. Lattice
defects including dislocations can be generated in the high
stress field region caused by an interaction between the
strained SiGe layer and the N-channel device structure. When
the dislocations are formed, fast moving transition metal
atoms such as Cu will be attracted to the dislocation site
and aggregate into nanodefects, known as "silicides". |
|
|
|
In the next few years and beyond, the nanoelectronic
manufacturing process will face major challenges in both
material and processing issues as fundamental limitations
are being approached. According
to the International Technology Roadmap for Semiconductors (ITRS),
an assessment of the semiconductor industry's technology
requirements says that a replacement material for silicon
with higher mobility will be necessary by the year 2010 [5].
We believe that advanced organic materials and semiconducting
carbon nanotubes
could pose a major threat to silicon,
if the fundamental problem with regards to silicon
nanodefects is not soon addressed.
REFERENCES
[1]
Nature, Issue October
18, 2001.
[2] EE Times,
Online December 17, 2003.
[3]
Hiroyuki
Fujimori, et al, Journal
of Crystal Growth 237–239 (2002) 338–344.
[4] Nikkei
Electronics Asia, Issue September 2002.
[5] Intel's Report, Online , 90 Nanometer: The
World's Most Advanced Chip Making Process.
[6] UM, Press Release #484, December 3, 2003.
Page
1 of 1
|
|
|
|
About the Author: Dr. Ed Wijaranakula is presently the
Chief Technical Officer at Infotix Systems, Inc. Prior to Infotix Systems, he has worked with
Intel, Hewlett-Packard, Micron, Motorola and Texas Instruments and has held senior as well
as managerial positions in semiconductor manufacturing companies. He has published over 80
technical papers and holds more than 12 U.S. and foreign patents. His portfolio holds long positions
or controls in IFX, INTC, LU and TXN.
|
|
|
|
