Information Technology Reference
In-Depth Information
physicist by the Nobel Prize committee when he was awarded the Nobel Prize
for physics in 2000 for his invention of the IC ( B.7.4 ).
The beginnings of Silicon Valley
After his invention of the junction transistor, Shockley had a falling out
with both Bardeen and Brattain. As a result, Bardeen left Bell Labs in 1951 to be
a professor at the University of Illinois at Urbana-Champaign. In 1972, Bardeen
won his second Nobel Prize for his role in developing the “BCS” (Bardeen,
Cooper, and Schrieffer) theory of superconductivity, the only physicist to have
been awarded two Nobel Prizes in physics. Shockley took leave of absence from
Bell Labs in 1953 and, with the help of a Caltech friend, Arnold Beckman, set
up the Shockley Semiconductor Laboratory, a division of Beckman Instruments
in Mountain View, California, in 1955 ( Fig. 7.8 ). Mountain View was near Palo
Alto, Shockley's hometown and the location of Stanford University. Three of
the first recruits to Shockley's company were physicists Robert Noyce and Jean
Hoerni and chemist Gordon Moore. Sadly, Shockley was a terrible people man-
ager and eventually alienated most of his employees. By the summer of 1957,
Noyce, Hoerni, Moore, and five others - the “Traitorous Eight” - decided to
leave Shockley and set up a company by themselves ( B.7.5 ). With financing
from the Fairchild Camera and Instrument Corporation, they formed Fairchild
Semiconductor. While their building was under construction, their temporary
workplace was a large garage in Palo Alto - now a tradition for Silicon Valley
start-ups!
Two key innovations were needed to make the mass production of power-
ful and robust ICs a reality. At the time, the state of the art in transistors was
the silicon mesa transistor , which had a tiny round plateau or “mesa” of silicon
set on top of a base, also made of silicon ( Fig. 7.9 ). Because the contacts on the
mesa structure were exposed, these transistors were easily contaminated or
damaged. Soon after Fairchild Semiconductor was established, the Swiss physi-
cist Jean Hoerni ( B.7.6 ) came up with a brilliant innovation, which now under-
pins all modern ICs. His idea was the planar transistor , where the mesa is now
fully embedded into the silicon wafer, resulting in a completely flat transistor
( Fig. 7.10 ). Hoerni also coated his device with a layer of silicon dioxide, which
insulated and protected the transistor's contacts. The remaining obstacle to
mass production was a method to electrically isolate the components in the
silicon. This was solved in late 1958 by the Czech-born physicist Kurt Lehovec,
who worked for the Sprague Electric Company in Massachusetts. He had heard
of Kilby's IC and realized the importance of isolating the different components
in the silicon. His solution was very simple: he proposed inserting back-to-back
p-n junctions, or diodes, between the transistors in the silicon so that no cur-
rent could flow in either direction. All these ideas came together in January
1959 when Noyce ( B.7.7 ) developed a process for manufacturing ICs using
Hoerni's planar transistors and Lehovec's p-n junctions. As Noyce said later:
Fig. 7.8. Memorial sign for the site of
the original Shockley Semiconductor
Laboratory. In later life, Shockley
became a controversial figure for his
views on race and genetics. Despite this
controversy it seems regrettable that
an earlier plaque with Shockley's name
on it has been replaced with this more
“politically correct” version.
B.7.5. A photograph of the
Traitorous Eight - the founders of
Fairchild Semiconductor who left
Shockley's original Silicon Valley
start-up. From left to right, they are
Gordon Moore, Sheldon Roberts,
Eugene Kleiner, Robert Noyce, Victor
Gingrich, Julius Blank, Jean Hoerni,
and Jay Last.
When this [the planar process] was accomplished, we had a silicon surface
covered with one of the best insulators known to man, so you could etch
holes through to make contact with the underlying silicon. Obviously, then,
 
Search WWH ::




Custom Search