December 23, 1947 is a day for the history books. On this day, John Bardeen and Walter Brattain presented to their colleague William Shockley at AT&T’s Bell Laboratories a small hand-made contraption: the first transistor. The importance of their invention was immediately clear to the three. This could amplify electrical signals without sensitive and power-guzzling vacuum tubes. The foundation for one of the greatest technological upheavals in human history was laid.
To understand how the transistor works, it is worth taking a look at its predecessor, the triode. Once heated – hence the gentle glow – electrons flow from the cathode to the anode in the evacuated glass bulb. They have to pass through a grid that is negatively charged. The higher this negative charge, the more the electrons, which are also negatively charged, are slowed down and correspondingly fewer pass through the grid. The electrons that have made it through the grid are accelerated by it to the anode. Only a weak electric field that can control a large current is required for regulation. The triode is thus a controllable flow control valve. However, this has its limits: 100 MHz is the limit, the triode cannot handle higher frequencies, and the amplification factor is also low.
First test run with a shaky construct
The invention of the transistor was actually long overdue when Bardeen, Brattain and Shockley created a first working prototype at Bell Laboratories in the USA. On December 23, 1947, the shaky construct completed its test run in an oscillator. Mataré’s team only succeeded eight months later. All the credit therefore went to the US colleagues, they were awarded the Nobel Prize in Physics in 1956. The German team’s designation “Transistron” also failed to gain acceptance. The similar made-up word “transistor”, made up of “transfer” and “resistor (resistance)”, which describes the function of the controllable resistance well, found more approval.
Bell Labs’ first transistor was a bipolar transistor, a design that was dominant until the late 1960s. Only then did the more powerful field effect transistors gain the upper hand – although the Lilienfeld and Heil patents originally described such field effect transistors. But their production was only economically possible in the 1960s.
How the transistor works
Back to the classic, the bipolar transistor. Then as now, it had three connections: emitter, base and collector. Inside they are connected to three semiconductor materials that are stacked on top of each other. There is a voltage between the emitter and the base, with the positive pole at the base. If the voltage exceeds a certain threshold – around 0.7 volts for a silicon transistor – the resistance drops and the transistor is switched to conduction. From the emitter, the electrons then flow into the base, where some of them are trapped. However, because the layer of the base is very thin, only a few electrons – less than one percent – get stuck there, the majority literally floods the base and flows on to the collector. On the other hand, if the voltage between emitter and base is below the threshold value, no current flows into the base and the resistance between emitter and collector becomes infinite, no current flows there either: the transistor is blocked.
The transistor then works as a perfect switch and thus generates the states 1 and 0 in digital components. On the other hand, if the voltage between the emitter and the base is varied continuously, this leads to a large continuous current change in the collector. The analogy to the triode as a controllable flow control valve comes to mind here: the base corresponds to the grid.
But if you delve into the physical principles, the analogy quickly ends. While an electric field more or less slows down the electrons in the triode, it is a semiconductor effect in the transistor. Semiconductors are substances that are neither an insulator, such as glass or plastic, nor a perfect conductor, such as metals. They lie in between, where exactly can be set by doping. For this purpose, these substances are specifically “contaminated” with foreign atoms, whereby there are two types. Atoms with an excess of electrons, e.g. phosphorus, get n-conductors, atoms with an excess of holes, e.g. boron, p-conductor. Excuse me: holes? These are atoms with a lack of electrons into which electrons can fall, causing the holes to move (virtually) in the opposite direction. The base of the bipolar transistor consists of a thin, p-doped layer that captures some of the electrons from the emitter, resulting in the small current flow described above. In contrast, the emitter and collector are n-doped.
But that is not set in stone. In addition to the npn transistor, there are also pnp variants that have an inverse structure and where the current flows in the opposite direction. In fact, the first transistors were pnp types made from an n-doped semiconductor chip into which p-dopants were introduced by diffusion on both sides, thus creating emitter and collector. Modern transistors have a much more complex structure, they consist of many three-dimensionally arranged layers with different doping densities.
Transistor concept – an epochal Christmas present
The first transistor had nothing in common with this. In his experiment, Brattain cut off the tip of a gold-coated polystyrene wedge, creating two closely spaced gold contacts. He pressed this wedge onto a germanium crystal. After the test on December 23, 1947, the next day he described in his laboratory notebook the first semiconductor circuit that amplified human speech by a factor of 18. On December 24th, the team presented their discovery – and gave their employer an epochal Christmas present.
The three inventors received the 1956 Nobel Prize in Physics. The ceremony in Stockholm was the last time they met. They were at odds: Brattain and Bardeen claimed the technology for themselves, and Shockley suspected treachery and founded his own company.
(jle)
