|
Semiconductor device
An electronic device whose essential characteristics are
governed by the flow of charge carriers
within a semiconductor.
History
The oldest ancestor of semiconductor devices was the
crystal detector, used in early wireless radios. This device
(patented by a German scientist, Ferdinand Braun, in 1899)
was made of a single metal wire (fondly called a "cat's
whisker") touching against a semiconductor crystal. The
result was a "rectifying
diode" (so called because it has
two terminals), which lets current
through easily one way, but hinders flow the other way. By
1930, though, vacuum-tube diodes
had all but replaced the smaller but much quirkier crystal
detector. The crystal and "cat's whisker" were left to
languish as a kids' toy in the form of "crystal radios."
The development of radar during World War II did much to
revive the fortunes of crystal detectors (and, as a result,
that of semiconductor devices) -- although temperamental,
crystals were better than vacuum-tube diodes
at rectifying the high
frequencies used by radar. So, during the war, much effort
was put into improving the semiconductors,
mostly silicon and germanium, used in crystal detectors. At
about the same time, Russell
Ohl at Bell Laboratories discovered that these materials
could be "doped" with small
amounts of foreign atoms to create interesting new
properties.
Depending on the selection of impurities (often called
dopants) added, semiconductor material of two
electrically-different types can be created -- one that is
electron-rich (called N-type,
where N stands for Negative), or one that is
electron-poor (called P-type,
where P stands for Positive). Most of the "magic" of
semiconductor devices occurs at the boundary between
P-type and N-type
semiconductor material -- such a boundary is called a
P-N junction. Ohl and his
colleagues found that such a P-N
junction made an effective diode.
For BEAMbots, we'll be concerned with two broad types of
semiconductor devices:
|