P-Type Doping

P-type doping adds Group III impurity atoms to a semiconductor crystal, creating an excess of holes — mobile positive charge carriers — that form the p-side of PN junctions and the base region of NPN transistors.

Why This Matters

P-type material is one of the two essential ingredients of every semiconductor device. Without it, there is no PN junction — the boundary between n-type and p-type is where all semiconductor action takes place. The base of an NPN transistor is p-type material. The entire body of a PNP transistor is structured with p-type emitter and collector around an n-type base.

Understanding p-type doping tells you what a hole is and why holes can carry current, what materials to use as acceptors, how much dopant to add, and how to verify the result. This knowledge is essential for anyone attempting semiconductor fabrication, and it clarifies the physics behind why the base of a transistor controls collector current despite carrying very little current itself.

The Physics of P-Type Doping

Silicon has four valence electrons. Each silicon atom bonds to four neighbors, satisfying all four bonding positions. Pure silicon conducts poorly because there are no free charges available to carry current at room temperature.

Group III elements — boron, gallium, indium, aluminum — have only three valence electrons. When such an atom substitutes for a silicon atom in the crystal lattice, it can only form three bonds with its four neighbors. The fourth bonding position remains unsatisfied — an empty orbital that can accept an electron. This empty site is a “hole.”

What a hole actually is: It is not the absence of matter — it is an electron vacancy that can migrate through the crystal as neighboring electrons jump into it. When an electron from one bond jumps to fill a hole, it leaves a new hole behind. The hole appears to move in the direction opposite to the electron. Because the crystal lattice has net positive charge around each hole (the acceptor atom with three electrons instead of four creates a local electron deficit), the hole behaves as a mobile positive charge carrier.

At room temperature, virtually all acceptor atoms are ionized — each one creates one hole. The material is electrically neutral overall (negative acceptor ion + positive hole), but rich in mobile positive carriers.

Acceptor Elements

DopantAtomic NumberIonization Energy in SiNatural Source
Boron (B)50.045 eVBorax (sodium tetraborate), colemanite, ulexite
Aluminum (Al)130.057 eVBauxite, common clay (Al₂O₃·2SiO₂·2H₂O)
Gallium (Ga)310.065 eVSphalerite (zinc ore), bauxite trace
Indium (In)490.160 eVSphalerite, zinc concentrates

Boron is by far the most common p-type dopant in silicon because:

  • Ionization energy of 0.045 eV — essentially fully ionized at room temperature
  • Smallest of the acceptors — least lattice strain
  • Available as volatile compounds (boron tribromide BBr₃, boron trichloride BCl₃, diborane B₂H₆) suitable for diffusion doping
  • Borax is a common mineral and has been traded for centuries

Indium is preferred for germanium because:

  • Melting point 157°C — melts and alloys easily into germanium at 500–550°C
  • Well-matched diffusivity in germanium
  • Used in the original Bardeen-Brattain-Shockley transistors (1947)
  • Indium pellets can be directly alloyed to form p-type regions by hand

Aluminum is tempting because it is the most common metal in Earth’s crust, but it oxidizes so readily that controlled doping is difficult without inert-atmosphere furnaces.

Doping Concentration and Electrical Properties

Doping LevelBoron Atoms/cm³Resistivity (Ω·cm)Typical Use
Very light10¹³–10¹⁴10–100High-resistivity substrates
Light10¹⁵–10¹⁶0.1–10Transistor base regions
Medium10¹⁶–10¹⁷0.01–0.1Diode anodes
Heavy (p+)10¹⁸–10²⁰<0.01Ohmic contacts, PNP emitter

In an NPN transistor the base is lightly p-type — low hole concentration means low recombination with injected electrons, giving high current gain. If the base were heavily p-doped, most electrons from the emitter would recombine before reaching the collector, and gain would be near 1.

Methods of P-Type Doping

Melt Doping with Boron

Add boron during Czochralski crystal growth:

  1. Prepare master alloy: Dissolve known mass of boron in silicon melt at 1:10,000 mass ratio. Example: 10 mg boron in 100 g silicon = 0.01% by weight. This gives approximately 10¹⁸ boron atoms/cm³.
  2. Dilute for final doping: Add chips of this master alloy to your main silicon melt. For 10¹⁶ /cm³ target, use 1 g master alloy per 100 g silicon.
  3. Pull crystal: Boron distributes fairly uniformly along the crystal (segregation coefficient close to 1, unlike phosphorus).

Boron’s segregation coefficient in silicon is 0.8, meaning it distributes nearly uniformly — the first-pulled and last-pulled sections differ by only a factor of a few, much better than phosphorus (k = 0.35).

Diffusion Doping with Boron

Introduce boron into a silicon wafer surface:

Using boron tribromide (BBr₃) vapor:

  1. Place silicon wafer in tube furnace at 950–1,000°C
  2. Pass nitrogen through liquid BBr₃ (boiling point 91°C) to carry vapor into tube
  3. BBr₃ decomposes on the silicon surface, depositing a boron-rich glass layer
  4. Boron diffuses from the glass into the silicon

Alternatively, solid boron oxide (B₂O₃) source:

  1. Mix powdered B₂O₃ with powdered silicon at ~10:1 ratio by weight
  2. Place this mixture near (not touching) the silicon wafer in the furnace
  3. At 1,000°C, B₂O₃ vaporizes and lands on the wafer surface
  4. Boron diffuses from the surface layer into the wafer

Diffusion depth for boron in silicon:

Temperature30 min2 hr8 hr
950°C~0.05 µm~0.1 µm~0.2 µm
1000°C~0.15 µm~0.4 µm~0.9 µm
1100°C~0.6 µm~1.5 µm~3 µm

Boron diffuses slightly more slowly than phosphorus in silicon, which is useful — boron p-type regions stay where you put them.

Alloying Indium into Germanium (Historical Method)

This is the most accessible p-type doping method for a rebuilding scenario:

  1. Start with n-type germanium wafer (phosphorus or antimony doped)
  2. Cut small indium pellets: ~0.5–1 mm diameter, or flatten indium wire
  3. Clean wafer surface with dilute acid, then rinse with water
  4. Place indium pellet at desired junction location
  5. Heat to 520–540°C in nitrogen atmosphere for 3–5 minutes
  6. Indium (melting point 157°C) melts and partially dissolves into germanium
  7. Cool slowly: 1°C per minute down to 200°C, then faster
  8. The recrystallized region is heavily p-type (indium concentration ~10¹⁹/cm³)
  9. Boundary between indium-alloyed region and n-type bulk = PN junction

Verifying P-Type Doping

Hot-Point Probe Test

  1. Heat a metal probe to ~200°C (glowing wire tip or soldering iron)
  2. Touch hot probe and cold probe to semiconductor surface, 1 cm apart
  3. Connect voltmeter: positive to hot probe, negative to cold probe
  4. P-type result: voltmeter reads positive (hot probe shows higher voltage)
  5. Reason: holes (the majority carriers) diffuse from hot end to cold end, leaving a net positive charge at the hot end

This is the opposite polarity from n-type material — the test definitively identifies carrier type.

Diode Test on a Formed Junction

After forming a PN junction by alloying indium onto n-type germanium:

  • Forward bias (p+ to +, n to −): resistance should be 50–500 Ω
  • Reverse bias (p to −, n to +): resistance should be >100 kΩ
  • Ratio >1000:1 indicates a clean junction

Summary

P-Type Doping — At a Glance

  • P-type doping adds Group III atoms (boron, gallium, indium, aluminum) to semiconductor crystal
  • Each acceptor atom creates one hole — a mobile positive charge carrier
  • Boron is standard for silicon; indium is easiest for germanium (melts and alloys at 520°C)
  • Hole concentration equals dopant concentration for typical doping levels
  • NPN transistor base is lightly p-doped (~10¹⁶/cm³) for low recombination and high gain
  • Melt doping: add boron to silicon melt during crystal growth (boron segregation coefficient ~0.8)
  • Diffusion doping: B₂O₃ or BBr₃ vapor at 950–1,100°C for surface p-type layers
  • Hot-point probe: p-type shows positive voltage at the hot probe