Skip to main content

PN junction formation and built-in potential

Short answer

When p and n regions are joined, carrier diffusion creates a depletion layer with ionized acceptors and donors, setting up a built-in field. At equilibrium, drift balances diffusion. Integrating the hole current equation yields built-in potential V0=(kT/e) ln(Na Nd/ni²).

Explain how PN junction is formed when n-type and p-type semiconductor are brought together. Derive the relation of built-in-potential of a PN junction.

Formation of a pn junction

A pn junction is formed when one region of a single crystal is made p-type and the adjacent region n-type (e.g. by diffusion or ion implantation). On contact:

  1. Diffusion: the p side has many holes and the n side many electrons. Holes diffuse into the n side and electrons into the p side, where they recombine with majority carriers.
  2. Space charge: near the junction the p side loses holes and is left with fixed negative acceptor ions Na−N_a^-; the n side is left with fixed positive donor ions Nd+N_d^+. This carrier-free layer is the depletion region of width W0=Wp+WnW_0 = W_p + W_n.
  3. Built-in field: the ions set up a field E0\mathcal{E}_0 from n to p. It drives a drift current opposite to the diffusion current.
  4. Equilibrium: diffusion continues until drift exactly balances it. The net current is zero, the Fermi level is flat across the device, and a potential difference V0V_0 (the built-in potential) exists between the n and p sides.
    p              W0             n
 o o o o |- - - - |+ + + + | * * * *
 o o o o |- - - - |+ + + + | * * * *
         |<- Wp ->|<- Wn ->|
          E0 <---------------
 V:  0 ______/^^^^^^^^^^^^ V0

Derivation of built-in potential

Consider holes. In equilibrium the hole drift current balances the hole diffusion current:

epμhE−eDhdpdx=0ep\mu_h\mathcal{E} - eD_h\frac{dp}{dx} = 0

With E=−dV/dx\mathcal{E} = -dV/dx and the Einstein relation Dh/μh=kT/eD_h/\mu_h = kT/e:

−pdVdx=kTedpdx  ⇒  dV=−kTedpp-p\frac{dV}{dx} = \frac{kT}{e}\frac{dp}{dx}\;\Rightarrow\; dV = -\frac{kT}{e}\frac{dp}{p}

Integrate from the p side (V=0V = 0, p=pp0p = p_{p0}) to the n side (V=V0V = V_0, p=pn0p = p_{n0}):

V0=kTeln⁡pp0pn0V_0 = \frac{kT}{e}\ln\frac{p_{p0}}{p_{n0}}

With pp0=Nap_{p0} = N_a and pn0=ni2/Ndp_{n0} = n_i^2/N_d:

V0=kTeln⁡NaNdni2V_0 = \frac{kT}{e}\ln\frac{N_aN_d}{n_i^2}

Equivalently, pn0/pp0=e−eV0/kTp_{n0}/p_{p0} = e^{-eV_0/kT}: the Boltzmann relation across the barrier.

Example: Si, Na=1018N_a = 10^{18}, Nd=1016 cm−3N_d = 10^{16}\ \text{cm}^{-3}, 300 K: V0=0.02585ln⁡(1034/2.1×1020)≈0.81V_0 = 0.02585\ln(10^{34}/2.1\times10^{20}) \approx 0.81 V.

✓ Answer checked by Claude Opus 5.5 (high effort)

All Electrical Engineering Material chapter 6 questions →