Join Barron Stone for an in-depth discussion in this video Semiconductor materials, part of Electronics Foundations: Semiconductor Devices.
- [Instructor] The materials used in electronic devices are commonly categorized as being either an electrical conductor or an electrical insulator. For example, metals like gold and copper are used as conductors because they allow electrons to move freely amongst their atoms. On the other hand, in materials like glass and rubber, electrons are bound more tightly to the atoms and have a harder time moving around. So those materials make good insulators. In this course, I'll be focusing on devices that use a third category of materials that exist somewhere between conductors and insulators, called semiconductors.
Semiconductor materials are able to conduct electricity, but only partly. They can conduct current better than an insulator, but not as well as a conductor. The most commonly used semiconductor material is silicon, which is used to manufacture everything from computer chips to solar panels. Silicon is popular because it's relatively cheap, it has good electrical properties and it's easy for manufacturers to fabricate into electronic devices.
To understand why silicon acts as a semiconductor, we need to look at its atomic structure, the number of electrons in a silicon atom, how those electrons are organized and how silicon atoms are bound together. An atom of silicon has four electrons that it can use to bond with other nearby atoms. And those are called its valence electrons. A silicon atom will form bonds with the four atoms closest to it by sharing its valence electrons with them.
And each of those four atoms will, in turn, form bonds with their nearest neighbors by sharing their own four valence electrons. And so on, which forms a crystal structure of silicon atoms. Since all of those valence electrons are tied up in forming bonds with adjacent atoms, they're not very mobile. They can't move around easily from atom to atom to carry electrical current. So on its own, although pure silicon qualifies as a semiconductor, electrically, it's not all that interesting.
To make silicon useful for electronic devices, a process called doping is used to inject other elements into the silicon structure to give it more mobile charges. Semiconductor materials like silicon can be doped in one of two ways, called n-type and p-type. The letters N and P refer to the sign of the charge that can move around within those doped materials. So in an n-type material, it's the negatively charged electrons which can easily move around and in a p-type material, it's the positively charged holes that move around.
Those holes represent an absence of electrons. To make an n-type semiconductor, pure silicon is injected with a small amount of another element whose atoms have one more valence electron to use for bonding than silicon atoms. Phosphorous is commonly used for that purpose because phosphorous atoms have five valence electrons and they're about the same size as silicon atoms, so it's easy for them to replace some of the silicon atoms in the crystal structure.
The injected phosphorous atoms use some of their electrons to bond with their four nearest neighbors, like a silicon atom would, but since the phosphorous atom has one more valence electron, that spare electron doesn't get tied up in a bond. That extra electron from the phosphorous atom can move around amongst the silicon atoms relatively easily, which gives the n-type doped silicon more mobile negative charges to use for conducting current. To help me conceptualize that movement, I like to think of the atoms in a silicon crystal like stacks of coins next to each other.
In pure silicon, each of the stacks in this row has four coins in it, representing the four valence electrons in each atom. If I apply a voltage to the stacks by pushing on one end with my finger, I can create a current by moving coins, but it's difficult. Moving one coin requires the entire row to shift with it. Now if I dope this row of silicon by replacing one of the stacks with a phosphorous atom, which has five valence electrons, with that extra coin on the stack, I can easily create a current by sliding the fifth electron from the phosphorous across the other stacks.
It's not tightly bound to its neighboring atoms. To make a p-type semiconductor, the pure silicon structure is injected with another type of element that has one less electron available to use for bonding, like boron, which only has three valence electrons. That lack of an electron creates a hole amongst the bonds that other electrons can jump into to fill, making it easier for electrons to move around. So when a voltage is applied to the p-type material, which pushes on its electrons, as those electrons move through the material by shifting into open holes, they leave behind a hole in the place they just came from.
Even though it's actually the negatively charged electrons that are moving from atom to atom to carry current through a p-type semiconductor, we consider the holes that are moving in the opposite direction to represent positive charge and that flows in the direction of conventional current. I can conceptualize that more easily using my stack of coins by removing a coin from one of the silicon stacks to replace it with a boron atom, which only has three valence electrons. Now because of the hole created by the boron, it's easy to move around the top layer of the electrons by shifting them into that open hole.
Shifting electrons towards the left to fill the hole causes the open hole to effectively drift towards the right, which is the direction of conventional current through this stack of coins. One common misconception about doped semiconductor materials is that n-type materials are negatively charged and p-type materials are positively charged. In actuality, n-type and p-type semiconductors are both neutral because the individual atoms of silicon and phosphorous and boron that make up those materials are each neutrally charged because they have the same number of electrons and protons inside of them.
The terms n-type and p-type just refer to the types of charge that easily move through the material, either negatively charged electrons or positively charged holes. So I've taken a semiconductor like silicon, which, on its own, can partially conduct electricity and have injected it with boron or phosphorous to give it more mobile charges so it can conduct electricity better. Why does that matter? If I just wanted a material that could conduct electricity well, it would be a lot easier for me to use a metal like copper or gold, than to go through all this trouble to dope a bunch of silicon.
On their own, n-type and p-type silicon are not all that exciting. They're just materials that conduct well. The magic really happens when they're used together. Semiconductor components like diodes and transistors have structures that contain both n-type and p-type materials. Certain combinations of those materials produce components that behave in different ways when certain voltages are applied to them and can be used to control and direct the flow of current through devices to create everything from power amplifiers to digital computers.
- Semiconductor materials
- Diode applications
- Rectifying a signal
- Detecting the signal peak
- Protecting against large signals, reverse current, and flyback voltage
- Special purpose zener diodes, Schottky diodes, and photodiodes
- NPN and PNP bipolar junction transistors
- Using a BJT as a switch
- Field effect transistors
- Differences between BJTs and MOSFETs
- Operational amplifiers
- Op-amp applications
- Comparing signals
- Buffering signals
- Amplifying signals
- Filtering signals
- Combining signals