Re: How do logic gates work

There are several different technologies that can be used to make logic gates, including bipolar technologies like ECL (emitter coupled logic) and TTL (transistor-transistor logic) - but, these days, most of the logic gates are built using MOS transistors.

MOS stands for Metal-Oxide-Semiconductor, which is kind of the "sandwich" like structure that you would see if you were to cut through an MOS transistor (cutting through the transistor is called "taking a cross section").

The "Metal" in MOS refers to the conductor on top, the "Oxide" in MOS means silicon dioxide (glass) underneath, which is an insulator. The "Semiconductor" in MOS refers to the silicon underneath the silicon dioxide - which can conduct if there are charges to carry electricity available in the silicon.

The Semiconductor is doped one type, say "P" , directly underneath the "Oxide", but is doped the opposite type, say "N", to the sides. This may be easier to visualize in the figure below.

The cross section of an MOS transistor looks like this:

"N" means that the semiconductor, silicon, contains a small amount of a chemical (phosphorus or arsenic or antimony) that allows that part of the semiconductor to have lots of free electrons available. The "P" region (in blue in the figure) of the semiconductor contains a different chemical (boron) that allows that part of the semiconductor to have almost free electrons, but instead have "holes" that represent places where electrons SHOULD be.

When a low or "zero" voltage is applied to the "Metal" (the GATE) of the MOS transistor, no electricity flows between the two "N" regions in light red in the figure.

When a high or "one" voltage is applied to the GATE of the MOS transistor, well -- opposites attract....the high, positive voltage on the Gate attracts negatively charged electrons underneath the GATE. However, since the OXIDE is an insulator, the electrons can't get through the OXIDE, and just pile underneath -- forming a layer through which electricity can be conducted from one "N" region (rich with electrons), through the layer of electrons that has been attracted underneath the GATE (this layer is called "the N-Channel"), to the other "N" region (also rich with electrons). This is illustrated below:

An MOS transistor that conducts with an "N-Channel", like above, is called an N-Channel MOS transistor, abbreviated NMOS. If we change all the "N' areas to "P" and all the "P" areas to "N" in the transistor, we have a P- Channel transistor, abbreviated PMOS.

If you connect NMOS and PMOS transistors together in the right way, you have the main type of logic circuits used today, CMOS logic gates.

Now, imagine that we are going to turn this cross-sectional view sideways, and put that cross sectional view over to the far right in the next few web pages. Click here to see a top view at the top, the cross sectional view to the right, and the circuit diagram underneath, for the simplest CMOS logic gate, an invertor.

Then, click here to see these views for another logic gate, called a NOR gate, which has two inputs.