Blog entry

What is a transistor?

By Wenjing Yan (CIC-Nanogune)

Transistors in electronic device are like synapses that connect neurons, if one were to put it in a more realistic picture. They are tiny little electronic components that you can find very easily in pretty much all electronic devices. You can get a feeling of how small it is just by imaging there are a few billions of them in a mobile or computer chip, and each of them is approximately 1/1000 of the thickness of a human hair.

To understand how transistor works, we have to introduce a fundamental particle, that is electron which carries a negative charge.  Its counterpart, a hole, carriers a unit of positive charge.

You can get excess electrons (holes) by deliberately put in an element that has more electrons (holes) than the host material, a process we call doping. Typically, the host material is Si, where there are no moving electrons because they are all tightly bonded in a chemical bond. Therefore, whatever excess you put into the system will be the conducting species. Like in the following schematics, you can have p doped material where holes are the conducting species, or n-doped, where electrons are the conducting species.

What the transistor does now is to regulate the flow of electrons/holes in the device.  The most simplest transistors are (1) a field effect transistor, which act like a switch that turns on and off, (2) a bipolar transistor, which does a bit more than just a switch. It can amplify the output voltage/current once triggered by a small input.

(1) A schematic of the field effect transistor looks like this:




It works in the following way.  For electrons to go from the source to the drain, it has to pass a channel , which is the green p doped area that has a sea of positive charges.  But they cannot go very far in to the p doped region, because of the potential that builds up in the interface between n type and p type material (the purple region, depletion width). This is what we call the OFF state. To create a conducting path, one can apply a positive gate voltage, so that the negative charges are attracted towards the surface in between the source and drain. Now the channel is conducting, and the device is in the ON state.

(2) A schematic of a bipolar transistor looks like this:


When the device is in the OFF state, electrons from the emitter could not diffuse to the base, because of the potential that build up in the interface (the purple area, depletion width).

To turn the device on, a small positive voltage is applied between the base and the emitter, so that the electrons have enough potential energy to drift to the base. A small part of the electrons don’t make it through the base and combines with holes, the rest of the electrons are swept through the base to the collector  by the potential difference build up in the purple depletion region on the right.  Therefore the small biase applied across emitter and base acts as a trigger to turn on/off the transistor.

Different types of transistor can work cooperatively to manage a task together. For example, they can do things like, if both A & B are on, then do this;  if either A or B is on, then do that. These then form the building blocks  for logic operation that does the computation.