2.2.4Doping & PN Junctions

Formation of a PN junction

1,924 words9 min readdifficulty · medium6 backlinks

WHAT is a PN junction?

Key players before contact:

Region Majority carriers Minority carriers Fixed ions (immobile)
p-type holes (pp) electrons negative acceptor ions NAN_A^-
n-type electrons (nn) holes positive donor ions ND+N_D^+

HOW it forms — the story in 4 steps

Step 1 — Diffusion (concentration wants to even out). There are way more electrons on the n-side and way more holes on the p-side. A concentration gradient exists.

Step 2 — Uncovering fixed ions (the depletion region). When an electron leaves the n-side, it leaves behind a fixed positive donor ion ND+N_D^+. When a hole leaves the p-side, it leaves behind a fixed negative acceptor ion NAN_A^-. Near the junction the mobile carriers are swept out.

Step 3 — Built-in electric field. The exposed ions form a dipole: ++ ions on the n-side, - ions on the p-side. This creates an internal electric field E\vec E pointing from n→p (from + charge to − charge).

Step 4 — Equilibrium (drift cancels diffusion). Diffusion keeps trying to push carriers across; the growing field pushes them back (drift). The region grows until: Jdiffusion=JdriftJnet=0J_{\text{diffusion}} = J_{\text{drift}} \quad\Rightarrow\quad J_{\text{net}} = 0 At this balance a stable built-in potential VbiV_{bi} exists across the depletion region and no external current flows.

Figure — Formation of a PN junction

Deriving the built-in potential VbiV_{bi} from first principles

Why this step? We use E=dVdxE = -\dfrac{dV}{dx} (field is the negative slope of potential) to swap field for potential: 0=qμnndVdx+qDndndx0 = -q\,\mu_n\, n\,\frac{dV}{dx} + q\, D_n \frac{dn}{dx}

Why this step? Divide by qnq\,n and rearrange to separate variables: dVdx=Dnμn1ndndx\frac{dV}{dx} = \frac{D_n}{\mu_n}\,\frac{1}{n}\frac{dn}{dx}

Why this step? Use the Einstein relation Dnμn=kTq\dfrac{D_n}{\mu_n} = \dfrac{kT}{q} (thermal voltage VT=kT/qV_T = kT/q), which links diffusion to thermal energy: dV=kTqdnndV = \frac{kT}{q}\,\frac{dn}{n}

Why this step? Integrate from the p-side (VpV_p, electron conc. npn_p) to the n-side (VnV_n, electron conc. nnn_n): VnVp=kTqln ⁣nnnpV_n - V_p = \frac{kT}{q}\ln\!\frac{n_n}{n_p}

So:


Depletion width intuition (Forecast-then-Verify)


Flashcards

What is a PN junction physically?
The boundary inside a single continuous crystal where doping changes from p-type to n-type.
Which carriers diffuse and in which direction when the junction forms?
Electrons diffuse n→p and holes diffuse p→n (high to low concentration).
What is the depletion region?
The zone near the junction emptied of mobile carriers, containing only fixed dopant ions.
Sign of exposed ions on each side?
Negative acceptor ions on the p-side, positive donor ions on the n-side.
Which way does the built-in electric field point?
From n-side to p-side (positive ions toward negative ions).
What condition defines equilibrium of the junction?
Drift current exactly cancels diffusion current, so net current = 0.
State the built-in potential formula.
Vbi=(kT/q)ln(NAND/ni2)V_{bi} = (kT/q)\ln(N_A N_D / n_i^2).
Typical VbiV_{bi} for silicon at room temperature?
About 0.6–0.7 V.
Einstein relation used in the derivation?
D/μ=kT/qD/\mu = kT/q (the thermal voltage).
Which side is the depletion region wider on?
The more lightly doped side (since NAxp=NDxnN_A x_p = N_D x_n).
Why doesn't VbiV_{bi} drive current in a shorted diode?
Contact potentials at the metal contacts cancel it; net loop EMF is zero.

Recall Feynman: explain to a 12-year-old

Imagine a crowded room (electrons on the n-side) next to an empty room with lots of open chairs (holes on the p-side). People rush from the crowded room into the empty one to sit down — that's diffusion, and each sitting-down is a recombination. But every person who leaves the crowded room leaves behind a "+" sticker on the wall, and every filled chair puts a "−" sticker on the other wall. Soon there's a wall of "+" stickers pulling the people back and "−" stickers pushing them back. Eventually the pushing balances the rushing, and everybody stops moving. That invisible wall of stickers is the depletion region, and the pull it makes is the built-in voltage.

Connections

  • Doping of Semiconductors — makes the p and n regions in the first place.
  • Intrinsic vs Extrinsic Semiconductors — source of nin_i and mass-action law np=ni2np=n_i^2.
  • Diffusion and Drift Currents — the two currents that balance at equilibrium.
  • PN Junction under Bias — what happens when you apply external voltage.
  • Diode I-V Characteristics — the exponential curve built on this barrier.
  • Einstein Relation — the D/μ=kT/qD/\mu = kT/q bridge.

Concept Map

placed next to

forms in single crystal

forms in single crystal

starts with

electrons n to p, holes p to n

uncovers fixed ions

exposes plus ND and minus NA

creates

drift opposes

grows until balances diffusion

is the finished

p-type region: holes

n-type region: electrons

PN junction

Diffusion of carriers

Recombination at boundary

Depletion region

Fixed ion dipole

Built-in E field n to p

Equilibrium barrier

Hinglish (regional understanding)

Intuition Hinglish mein samjho

Socho tumhare paas ek hi crystal hai jiska ek side p-type (holes zyada) aur doosra side n-type (free electrons zyada) doped hai. Jaise hi ye judte hain, concentration difference ki wajah se electrons n-side se p-side ki taraf aur holes p-side se n-side ki taraf diffuse karte hain — high se low concentration, bilkul ink ke paani me phailne jaisa. Junction ke paas ye milke recombine ho jaate hain.

Ab twist ye hai: jab electron n-side chhodta hai to peeche ek fixed positive donor ion reh jaata hai, aur jab hole p-side chhodta hai to peeche fixed negative acceptor ion reh jaata hai. Ye ions hil nahi sakte. Toh junction ke aas-paas ek depletion region ban jaata hai jisme sirf ye charged ions hote hain — n-side pe plus, p-side pe minus. Yahi charges ek built-in electric field banate hain jo n se p ki taraf point karta hai.

Ye field ab diffusion ko rokta hai — electrons ko wapas n ki taraf aur holes ko wapas p ki taraf dhakelta hai. Region tab tak badhta hai jab tak drift current = diffusion current na ho jaye. Is balance point pe net current zero hota hai aur ek built-in potential Vbi=(kT/q)ln(NAND/ni2)V_{bi} = (kT/q)\ln(N_A N_D/n_i^2) ban jaata hai — silicon ke liye ye lagbhag 0.7 V hota hai. Yahi wajah hai ki silicon diode 0.7 V ke aas-paas "on" hota hai. Yaad rakho: har side apne aap me neutral hai, charge sirf junction ke paas diffusion ke baad dikhta hai.

Go deeper — visual, from zero

Test yourself — Doping & PN Junctions

Connections