If you are asking what is doping in semiconductors, here is the short answer. It is the deliberate addition of a tiny amount of impurity atoms to pure silicon, which lets it conduct electricity in a controlled way.
This explainer belongs to the wider topic of semiconductor basics, so use that page if you want the surrounding concepts.
| Item | Value |
|---|---|
| Light doping | About one dopant atom per 100 million intrinsic atoms |
| Heavy doping | About one dopant atom per ten thousand atoms |
| Silicon band gap | About 1.12 eV |
| Boron binding energy in silicon | 0.045 eV |
| Intrinsic carrier concentration of silicon at 300 K | Roughly 1.08×10^10 cm^-3 |
These figures come from Wikipedia’s article on semiconductor doping. Values can change or differ between references, so check the source before you rely on them.
What doping does to pure silicon
Pure silicon is an intrinsic semiconductor, meaning nothing has been added on purpose. According to a RemNote study deck, such material holds very few free charge carriers (electrons and holes) at room temperature, which makes it a poor conductor. Wikipedia puts silicon’s intrinsic carrier concentration at roughly 1.08×10^10 cm^-3 at 300 kelvins, about room temperature.
Doping is the fix. In semiconductor production it is the intentional introduction of impurities into an undoped semiconductor to change its electrical, optical and structural properties, and the result is called an extrinsic semiconductor, as described on Wikipedia.
The key effect is on conduction. Small numbers of dopant atoms can change how well a semiconductor carries current, and in general more doping means more conductivity because there are more carriers, as Wikipedia explains.
A Purdue lecture on nanoHUB summarises the process in a few steps:
- Start with a pure crystal such as silicon.
- Replace a few of its atoms with atoms from a different column of the periodic table.
- The dopants ionize, producing electrons in the conduction band or holes in the valence band.
How donors and acceptors work
Dopants come in two families, and the family decides which carrier does most of the conducting. The RemNote study deck describes both:
- n-type (donor) doping: Group V elements with 5 valence electrons, such as phosphorus, arsenic or antimony. Silicon only needs 4 electrons to bond with its neighbours, so the fifth is weakly bound and easily released into the conduction band. The majority carriers are electrons.
- p-type (acceptor) doping: Group III elements with 3 valence electrons, such as boron, gallium or indium. The missing electron leaves an empty state called a hole, which neighbouring electrons can jump into. The majority carriers are holes.
In the most common method, substitutional doping, a Group V or Group III atom takes the place of a Group IV host atom in the lattice. The dopant must be similar in size to the atom it replaces, which is why these neighbouring columns suit silicon, according to the RemNote study deck.
For a closer side-by-side look, the guide on n-type vs p-type semiconductor compares the two materials directly.
The same study deck notes that donors and acceptors tend to neutralise each other when both are present. That is why practical devices are made by doping different regions with only one type of dopant. How those opposite regions behave where they meet is covered in the explainer on how does a pn junction work.
The common choices for silicon, per the same deck, are:
- Donors (n-type): phosphorus, arsenic and antimony.
- Acceptors (p-type): boron, aluminum and gallium.
How much dopant does silicon need?
Far less than you might expect. Wikipedia calls doping light at about one dopant atom per 100 million intrinsic atoms, and heavy at about one per ten thousand atoms.
Heavy doping is often written as n+ or p+, while minus signs mark lighter doping. Push it far enough and the material becomes a degenerate semiconductor, one that acts more like a conductor than a semiconductor.
For silicon, the same Wikipedia article says doping concentrations may range from 10^13 cm^-3 to 10^18 cm^-3. Above about 10^18 cm^-3 the material is considered degenerate at room temperature. Even those degenerate levels still mean a low concentration of impurities compared with the silicon itself.
Degenerate material has conductivity comparable to metals and is often used in integrated circuits as a replacement for metal, according to Wikipedia.
One rule keeps the picture tidy. At low doping and thermal equilibrium, the product of the electron and hole concentrations equals the square of the intrinsic carrier concentration, as Wikipedia shows. Adding donors therefore raises the electron count and pushes the hole count down, and the RemNote study deck sums it up as: add more donors, get more electrons.
Why does temperature matter for doped silicon?
Dopants work at room temperature because their binding energy is small. Wikipedia gives 0.045 eV for boron in silicon, against a silicon band gap of about 1.12 eV, and a course handout from MSOE lists 0.044 eV for phosphorus.
Because the energy gap is so small, room temperature is hot enough to thermally ionize practically all of the dopant atoms and create free charge carriers, as Wikipedia notes.
The behaviour changes with temperature. The nanoHUB lecture describes freeze-out, extrinsic and intrinsic regions:
- Low temperatures: semiconductors become insulators because the dopants are frozen out.
- Middle range: the extrinsic region, where the dopants set the carrier count.
- High temperatures: doped semiconductors become intrinsic again.
A short history of doping
The effects of impurities were known empirically long before anyone could control them, in devices such as crystal radio detectors and selenium rectifiers. Wikipedia notes that Shelford Bidwell in 1885 and Bernhard Gudden in 1930 each independently reported that semiconductor properties were due to the impurities they contained.
A doping process was formally developed by John Robert Woodyard at Sperry Gyroscope Company, and his US patent issued in 1950 describes adding tiny amounts of elements from the nitrogen column of the periodic table to germanium, according to Wikipedia. Similar work by Gordon K. Teal and Morgan Sparks at Bell Labs led to a US patent issued in 1953.
Frequently asked questions
Does doping turn silicon into a metal?
Show answer
Not in normal use. Only very heavy doping gives a degenerate semiconductor, which has conductivity comparable to metals, as Wikipedia explains.
Which elements are used to dope silicon?
Show answer
The RemNote study deck lists phosphorus, arsenic and antimony as donors, and boron, aluminum and gallium as acceptors.
Can one dopant act as both a donor and an acceptor?
Show answer
In some materials, yes. In gallium arsenide, silicon is an acceptor on an arsenic site and a donor on a gallium site, according to the nanoHUB lecture.
What do n+ and p+ mean?
Show answer
They mark heavy doping of n-type and p-type material. Wikipedia adds that superscript plus and minus signs denote relative doping concentration.
