Bill shows how a transistor works by examing a replica of the first one ever build: The Bardeen-Brattain point contact transistor.
Transcript This doesn't look like it would change the world, but this first transistor did just that. This replica is a bit larger than the original, but otherwise a pretty faithful representation. In 1947 Walter Brattain at Bell Lab fashioned it out of plastic stand, a chunk of germanium, a triangular piece of plastic with a layer of gold on each side, and a "spring" on top to press the gold at the apex of the triangle into the germanium - that's why its called a "point contact" transistor.
Here's what Bardeen and Brattain actually did with this device on December 23, 1947.
They hooked up a microphone to the left side, and an oscilloscope to the right side. As they spoke into the microphone they could see the voice signal being amplified. As Brattain wrote in his lab notebook "This circuit was actually spoken over and by switching the device in and out a distinct gain in speech level could be hear and seen on the scope ...."
Now, that amplification is still one of the main uses for a transistor. Think of your cell phone: It detects a low-power signal from the cell tower and then its circuity amplifies the signal until you can hear it.
But how does this klutzy looking contraption work its magic? They key lies in this chunk of semi-conducting germanium.
Recall that there are three ways we can classify material in terms of their ability to conduct electricity.
Conductors like metals that readily transport electricity with negative charge carriers. Insulators which refuse to allow current to flow. And a third class, which make possible the transistor: semiconductors. As the name implies they they conduct better than insulators, but not as well as conductors. But more importantly unlike metals semiconductors have two different ways to conduct electricity - effectively negative and positive charge carriers.
That property lies at the heart of a transistor. It allows an engineer to make a highly reliable device that allows current to flow in only one direction. We can make a "sandwich" of the two types of semiconductors. Here the negative charge carrier one on the left and the positive one on the right. The semiconductor sandwich allows electricity to flow. Reverse the battery and the current grinds to a halt.
Other devices can do this, for example, vacuum tubes, but they had many part, were unreliable, overheated, and were expensive to produce and likely impossible to miniaturize.
This seems a simple thing, yet it lies at the heart of our microelectronic revolution; and it is the key to the Brattain and Bardeen transistor - in fact to every transistor.
It's best to first think of their device as two separate one-way current valves. At the center of each is that piece of germanium, which makes electric contact with a piece of copper at its base. And at the top a spring presses a thin piece of gold foil onto the germanium.
Right where the gold touches there is a thin layer of the positive charge carrier semiconductor, below that the rest of the germanium is of the negative carrier type. Even though that top layer is very thin I've exaggerated it here to make the operation of the transistor clear.
On the left we attach a small battery with its positive terminal connected to the top layer: This allows current to flow. On the right we use a larger battery, but with its ends reversed. This creates a device where current doesn't flow. The magic happens when we put the two in contact. The current flows on the left. But, because the distance between these gold contacts is less than two-thousands of an inch, the positive charge carriers in the thin layer side are "injected" into or "stolen" by the right side - so now current flows there. The current on the right is controlled by that on the left.
It seems we've gone to great lengths to just make a tiny current flow - I mean the current flow on the right isn't too much bigger than on the left. But if we look carefully we can see why this revolutionized the world: We've created a signal amplifier. The key element is the battery. Recall that power is current times voltage. So, if we use a small battery on the left, and a large on on the right we have a device that amplifies any fluctuations in the current on left side - those fluctuations might be, for example, the output of the microphone that Brattain and Bardeen used on December 23, 1947.
Western Electric company manufactured these point-contact transistors in 1951, but they found it difficult to make them reliable, some even showing "sudden death." From a manufacturing viewpoint these transistor have a fatal flaw: They are three dimensional. As Brattain said "no good physicists likes to work with a complicated case if it can be reduced to one dimension." Within a year Western Electric, Raytheon, RCA, and General Electric offered the junction transistor - a superb one-dimensional device. I'm Bill Hammack, the engineer guy.