P-type doping with acceptor atoms (boron): Visual walkthrough

We will only use two physical facts, and we will earn both with a picture before touching algebra. Every symbol is defined the first time it appears.

Step 1 — Name the two things we are counting

WHAT. Inside the silicon crystal there are two kinds of mobile charge that can carry current. We give each a single letter:

WHY letters and not words. We are about to write down relationships between "how many electrons" and "how many holes." Sentences can't be added or squared; symbols can. So we compress each count into one letter.

PICTURE. Below, the crystal is a grid. Teal dots = free electrons ( $n$ of them). Orange rings = free holes ( $p$ of them, drawn as an empty circle = "missing electron"). Fixed boron ions sit as plum minus-signs — they are stuck, so they carry current for nobody.

Figure — P-type doping with acceptor atoms (boron)

Notice: in p-type there are many orange holes and very few teal electrons. That imbalance is exactly what we are going to compute.

Step 2 — Fact one: the see-saw law $n p = n_{i}^{2}$

WHAT. No matter how much boron we add, the product of the two counts is pinned to a fixed number: $electrons n \cdot holes p = a constant set by temperature n_{i}^{2}$

Here == $n_{i}$ == ("intrinsic concentration") is how many electrons and holes pure, undoped silicon has on its own — for silicon at room temperature, $n_{i} \approx 1.5 \times 1 0^{10} cm^{- 3}$ . (This is the baseline from Intrinsic Semiconductors.)

WHY is the product constant? Electron–hole pairs are constantly being created by heat (a rate that depends only on temperature and the band gap — not on how many carriers already exist) and destroyed by recombination (an electron meeting a hole — a rate proportional to $n \cdot p$ , because you need one of each). At equilibrium creation = destruction, so $n p$ locks to a fixed value. The full argument lives in Mass-Action Law and Carrier Statistics.

WHY the tool "product = constant" and not "sum = constant"? Because recombination needs one electron AND one hole to meet — the meeting rate multiplies the two counts. Multiplication, not addition, is what physics rewards here. That is precisely why the constraint is a product.

PICTURE. The see-saw: as we slide $p$ (holes) up along the curve, $n$ (electrons) slides down. Every point on the orange curve has the same product $n_{i}^{2}$ .

Step 3 — Fact two: the crystal is electrically neutral

WHAT. The whole block of silicon has zero net charge. Add up every charge inside and it must cancel.

Which charges are there? Four kinds:

mobile holes, + (+ 1) p + ionized donors, + (+ 1) N_{D}^{+} = mobile electrons, - (- 1) n + ionized acceptors, - (- 1) N_{A}^{-}

Every symbol, term by term:

$p$ = holes, each $+ 1$ → positive pile.
== $N_{D}^{+}$ == = number of ionized donor ions per $cm^{3}$ (phosphorus, from N-type doping with donor atoms (phosphorus)). Each donor that gave up an electron is left $+ 1$ . In pure p-type there are none: $N_{D} = 0$ .
$n$ = electrons, each $- 1$ → negative pile.
== $N_{A}^{-}$ == = number of ionized acceptor ions per $cm^{3}$ . Each boron that grabbed an electron becomes a fixed $B^{-}$ , charge $- 1$ .

WHY this equation. "Neutral" literally means positive charge counted = negative charge counted. Writing that sentence in symbols gives the balance above. This is a bookkeeping fact, nothing more — but it links the dopant count $N_{A}$ to the carrier counts $n$ and $p$ , which is exactly the bridge we need.

PICTURE. A balance scale: the positive pan holds holes; the negative pan holds electrons plus the fixed boron ions. The pans are level → net charge zero.

Step 4 — Specialise to pure p-type

WHAT. Set $N_{D} = 0$ (no phosphorus). At room temperature every shallow boron accepts its electron, so all acceptors are ionized: $N_{A}^{-} \approx N_{A}$ , where == $N_{A}$ == is simply how much boron we put in. The neutrality equation collapses to:

$p = n + N_{A}$

Term by term: holes $=$ electrons $+$ the doping level.

WHY this shape makes sense. Read it as: holes come from two sources — the ones boron deliberately created ( $N_{A}$ ) plus the ones that arise naturally with an electron partner ( $n$ , the thermal pairs). Boron's holes have no matching electron (their electron went to fill boron's bond and got stuck as $B^{-}$ ), which is why they show up as a pure surplus $N_{A}$ .

PICTURE. A stacked bar: the total hole bar ( $p$ ) equals the thermal-pair block ( $n$ ) sitting on top of the boron surplus block ( $N_{A}$ ).

Step 5 — Two equations, two unknowns → one quadratic

WHAT. We now have a system: $n p = n_{i}^{2} (Step 2) p = n + N_{A} (Step 4)$

Two equations, two unknowns ( $n$ and $p$ ). Standard move: eliminate one unknown. From the see-saw law, $n = n_{i}^{2} / p$ . Substitute into the neutrality equation:

$p = \frac{n _{i}^{2}}{p} + N_{A}$

Multiply through by $p$ (allowed since $p > 0$ , there is always at least a whisper of holes):

$p^{2} = n_{i}^{2} + N_{A} p ⟹ p^{2} - N_{A} p - n_{i}^{2} = 0$

WHY a quadratic and not a line? Because the see-saw law is multiplicative ( $n$ and $p$ multiply). When we substituted $n = n_{i}^{2} / p$ , the $1/ p$ turned into a $p^{2}$ after clearing the fraction. The curved see-saw is the source of the square.

PICTURE. Two curves in the $(p, n)$ plane: the orange see-saw $n = n_{i}^{2} / p$ and the teal straight line $n = p - N_{A}$ (from neutrality). Their crossing point is the true $(p, n)$ — the one pair that obeys both laws at once.

Step 6 — Solve the quadratic, pick the physical root

WHAT. The quadratic $p^{2} - N_{A} p - n_{i}^{2} = 0$ has the form $a x^{2} + b x + c = 0$ with $a = 1$ , $b = - N_{A}$ , $c = - n_{i}^{2}$ . The quadratic formula $x = \frac{- b \pm b ^{2} - 4 a c}{2 a}$ gives:

$p = \frac{N _{A} \pm N _{A}^{2} + 4 n _{i}^{2}}{2}$

Term by term:

$N_{A}$ = the doping surplus reappearing (that was $- b$ ).
$N_{A}^{2} + 4 n_{i}^{2}$ = the discriminant. Since $n_{i}^{2} > 0$ this root is always larger than $N_{A}$ .
The $2$ downstairs is the $2 a$ of the formula.

WHY the $+$ sign only. A concentration cannot be negative — you cannot have $- 3$ holes per cm³. With the $-$ sign the root exceeds $N_{A}$ , making the numerator negative, giving $p < 0$ . Unphysical → discard. Only the $+$ root survives:

PICTURE. The two algebraic roots on a number line: one positive (physical, kept in orange), one negative (unphysical, crossed out in grey). Only one crossing lives in the real crystal.

Step 7 — Case A: heavy doping ( $N_{A} ≫ n_{i}$ )

WHAT. In real chips $N_{A} \sim 1 0^{16} - 1 0^{18} cm^{- 3}$ , while $n_{i} \approx 1.5 \times 1 0^{10}$ . So $N_{A}^{2}$ utterly dwarfs $4 n_{i}^{2}$ . Inside the root, drop the tiny term: $N_{A}^{2} + 4 n_{i}^{2} \approx N_{A}^{2} = N_{A}$ Then $p \approx \frac{N _{A} + N _{A}}{2} = N_{A}, n = \frac{n _{i}^{2}}{p} \approx \frac{n _{i}^{2}}{N _{A}}$

Term by term: holes are pinned to the dopant ( $p \approx N_{A}$ ); electrons are crushed far below intrinsic ( $n \approx n_{i}^{2} / N_{A} ≪ n_{i}$ ).

WHY it's allowed. The correction we ignored is of order $(n_{i} / N_{A})^{2} \sim (1 0^{10} /1 0^{16})^{2} = 1 0^{- 12}$ — twelve zeros of smallness. Negligible.

PICTURE. On the see-saw, the crossing sits far out on the $p$ axis: $p$ nailed to $N_{A}$ , $n$ scraping the floor.

Step 8 — Case B: light doping ( $N_{A} \sim n_{i}$ ) and the intrinsic limit ( $N_{A} \to 0$ )

WHAT. When boron is sparse, $N_{A}$ is not huge compared to $n_{i}$ , and the shortcut $p \approx N_{A}$ fails. We must keep the full root.

The degenerate limit $N_{A} \to 0$ (pure silicon). Set doping to exactly zero: $p = \frac{0 + 0 + 4 n _{i}^{2}}{2} = \frac{2 n _{i}}{2} = n_{i}$ and $n = n_{i}^{2} / p = n_{i}$ too. So $n = p = n_{i}$ — exactly the intrinsic result. Our formula smoothly reduces to the undoped crystal when the boron vanishes. That is the sanity check that the whole derivation is trustworthy: no boron, no imbalance.

PICTURE. Three see-saw crossings side by side: (left) $N_{A} \to 0$ landing at $n = p = n_{i}$ ; (middle) $N_{A} \sim n_{i}$ with a mild tilt; (right) $N_{A} ≫ n_{i}$ far out. One formula, all three regimes.

The one-picture summary

Everything above compressed into a single flow: the two physical laws feed into a quadratic, the quadratic is solved, and the two doping regimes fall out as limits of that one solution.

Recall Feynman: the whole walkthrough in plain words

We wanted to know how crowded with holes a boron-doped silicon block is. We only needed two truths. First truth: electrons and holes are on a see-saw — multiply their counts and you always get the same fixed number $n_{i}^{2}$ , because heat keeps making pairs and pairs keep recombining at a rate that needs one of each. Second truth: the block has zero net charge, so all the positive holes must balance all the negative electrons plus the stuck negative boron ions. Writing "holes = electrons + boron" and "electrons × holes = fixed" gives two equations. We slide one into the other, and the see-saw's multiply-turns-into-a-square trick hands us a quadratic. Solving it (and throwing away the impossible negative-holes answer) gives one clean formula for $p$ . Then we look at it two ways: pour in lots of boron and it says "holes ≈ boron, electrons ≈ almost none"; pour in no boron and it politely reduces to pure silicon's $n = p = n_{i}$ . One formula, every case covered.

Connections

Intrinsic Semiconductors — the $N_{A} \to 0$ limit where $n = p = n_{i}$ (Step 8).
Mass-Action Law and Carrier Statistics — where the see-saw $n p = n_{i}^{2}$ comes from (Step 2).
N-type doping with donor atoms (phosphorus) — the mirror derivation with $N_{D}$ (Step 3).
Fermi Level in Doped Semiconductors — where these carrier counts push the Fermi level.
PN Junction Formation — what happens when this p-type block meets an n-type one.
Drift and Diffusion Currents — how the holes we just counted actually carry current.