Intuition The ONE core idea
A semiconductor's usefulness comes down to counting two things: how many mobile electrons sit in the upper band and how many empty spots ("holes") sit in the lower band. Every symbol on the parent page exists to answer just one question — how many seats are there, and what fraction are filled?
Before you can read the carrier-concentration equations, you must own every letter and squiggle in them. This page introduces each one from absolute zero : what it means in plain words, the picture it stands for, and why the topic can't live without it. Nothing here assumes you've seen band diagrams before.
We build the vocabulary in the order the physics needs it — energy first, then bands, then "seats", then "occupancy", then the counting itself. Come back here whenever a symbol on the parent note feels unearned.
E
E is just how much energy one electron has , measured in electron-volts (eV). Think of it as height on a vertical ladder : the higher up an electron sits, the more energy it carries.
Every diagram in this topic uses energy going up the page . That single choice is why we can talk about electrons "falling" and holes "rising" later — it's a real height axis, not a metaphor.
Worked example Figure 1 — the energy ladder
What it shows: a vertical arrow labelled Energy E (eV) , with a horizontal-rung "ladder" of allowed energy levels. The orange dot is one electron sitting on a low rung; the plum dot is that same electron promoted to a higher rung. Read it as: the only thing that changes between the two dots is height = energy, not left–right position. This is why every later diagram plots energy upward.
Intuition Why energy and not position?
In everyday physics we plot where things are. Here we plot how energetic they are, because whether an electron can move and conduct depends entirely on its energy level — not on where it is in the crystal. So the whole subject is drawn on an energy axis .
Electrons in a solid are not allowed to have any energy — only energies inside certain bands (allowed ranges), separated by a forbidden zone. For a semiconductor there are two bands that matter.
Definition Band edges and gap
E v = valence band edge — the top of the lower, normally-full band (the "main floor").
E c = conduction band edge — the bottom of the upper, normally-empty band (the "balcony").
E g = E c − E v = the band gap — the height of the forbidden step between them.
The subscripts are memory hooks: c for c onduction, v for v alence.
Worked example Figure 2 — bands and the forbidden gap
What it shows: the same energy axis. The teal filled block at the bottom is the valence band (full of electrons), ending at the teal line E v . The faint orange block on top is the conduction band (empty), starting at the orange line E c . The plum double-arrow between them measures the gap E g = E c − E v . Read it as: nothing is allowed inside the plum span — that's the forbidden zone an electron must jump over to conduct.
Intuition Why the gap matters
E g is the "jump height" an electron must clear to become mobile. A big gap (like diamond) means almost no electrons make it across — an insulator. A small gap (like silicon, E g ≈ 1.12 eV) means a few make it with room-temperature heat — a semiconductor. This one number controls n i . See Band gap Eg .
Common mistake "The gap is where electrons live."
Feels right because we draw the gap as a big empty region. Truth: the gap is forbidden — no allowed states exist there. Electrons live in the bands around it. The only special energy inside the gap is a bookkeeping marker, E F , coming next.
E F
E F is the energy at which a state has a 50% chance of being filled. Picture it as the water line on the energy ladder: states well below are "underwater" (filled), states well above are "in the air" (empty), and right at E F it's a coin-flip.
Worked example Figure 3 — the Fermi level as a water line
What it shows: the horizontal axis is probability filled f ( E ) (from 0 to 1), the vertical axis is energy. The plum S-curve is the occupancy: near 1 at the bottom (filled seats "underwater"), near 0 at the top (empty seats "in the air"). The dashed teal line marks E F , exactly where the plum curve crosses f = 0.5 . Read it as: E F is not a seat — it's the height of the water line that separates mostly-filled from mostly-empty.
Intuition Why we need a water line
We can't list all 1 0 22 electrons. Instead we track a single number , E F , that tells us where the filled/empty boundary sits. Raising E F (by adding donor atoms) floods more of the conduction band with electrons. So E F is the one dial that encodes doping. More in Fermi level position and Doping and charge neutrality .
Notice E F appears inside every carrier equation as a difference : E c − E F or E F − E v . That difference is literally "how far the seat is above/below the water line."
Definition The thermal energy
k T
k is Boltzmann's constant (a fixed conversion number, k ≈ 8.62 × 1 0 − 5 eV/K ) and T is absolute temperature in kelvin. Their product k T is the typical jiggle-energy that heat gives to one particle. At room temperature (300 K), k T ≈ 0.0259 eV.
k T appears in every exponent
Every energy in this topic is compared to k T : an energy gap "feels big" only relative to how much thermal kick is available. That's why formulas contain ratios like k T E c − E F — a pure number answering "how many thermal kicks tall is this step?" A step of 3 k T or more is "hard to climb"; that threshold is exactly what makes the Boltzmann approximation valid.
k T is a temperature."
No — k T has units of energy (eV), because k converts kelvin into eV. It's the energy currency heat pays in.
Definition Density of states
g ( E )
g ( E ) tells you how many available electron-seats exist per unit energy and per unit volume at height E . Its units are states ⋅ eV− 1 ⋅ cm− 3 (or per m3 ). Big g = a crowded shelf with many seats; small g = a sparse shelf.
Worked example Figure 4 — the square-root shape of
g c ( E )
What it shows: energy on the vertical axis, density of states g c ( E ) on the horizontal axis. Below the dashed orange line E c there are zero conduction seats (nothing is drawn). Above E c the orange curve grows like E − E c — wider means more seats. Read it as: the higher you climb above the band edge, the more available seats appear, but right at the edge there are essentially none.
The subscript picks the band: g c ( E ) counts seats in the conduction band, g v ( E ) in the valence band. The parent note derives its square-root shape, g c ( E ) ∝ E − E c . Full detail lives in Density of states .
Intuition Why "density" — and why per volume too
There are infinitely many possible energies, so you can't count seats at an exact energy — you count seats within a thin slice of energy. That's what "per unit energy" means: g ( E ) d E = seats between E and E + d E . And because a bigger crystal has proportionally more seats, we also divide by volume, so g is per unit volume . That is exactly why the final carrier counts n , p come out as concentrations (seats per cm3 ), not raw totals.
f ( E )
f ( E ) = 1 + e ( E − E F ) / k T 1
f ( E ) is the probability that a seat at energy E is occupied — a number between 0 and 1.
Read it against the water-line picture:
Far below E F : exponent is a big negative number, e negative → 0 , so f → 1 (seat filled).
At E F : exponent = 0 , f = 1 + 1 1 = 2 1 (coin-flip — this is why E F is the 50% line).
Far above E F : exponent is big positive, denominator huge, f → 0 (seat empty).
Intuition Why this exact shape and not a hard cutoff
If electrons ignored each other you'd expect a sharp "filled below, empty above" step. But two electrons can't share a seat (Pauli exclusion), and heat blurs the edge. f ( E ) is the smooth blur: sharp at low T , softened by k T . The "+ 1 " in the denominator is the fingerprint of that exclusion rule. Deep dive: Fermi-Dirac distribution .
1 − f ( E )
A hole is a missing electron, so the probability a valence seat is empty is 1 − f ( E ) . This is why the hole integral on the parent page uses 1 − f ( E ) instead of f ( E ) — same idea, mirror image.
+ 1 " can be dropped
For a conduction seat, E ≥ E c . If the water line E F sits well below the band edge — specifically E c − E F ≫ k T (a gap of at least ∼ 3 k T ) — then every conduction seat has E − E F ≫ k T , so e ( E − E F ) / k T is a huge number. Next to a huge number the "+ 1 " is negligible, and
f ( E ) = 1 + e ( E − E F ) / k T 1 ≈ e − ( E − E F ) / k T .
This simple exponential is the Boltzmann tail . Calling this the non-degenerate case (water line safely inside the gap) is exactly what lets the messy Fermi–Dirac integral collapse into the clean formulas in Section 9. If instead E F climbs into the band (heavy doping), the "+ 1 " matters and you must keep the full integral — see Intrinsic vs extrinsic semiconductors .
Definition Effective mass
m n ∗ (for electrons) and m p ∗ (for holes) are the apparent masses the carriers behave with inside the crystal , which differ from the free-electron mass because the atomic lattice pushes back.
Intuition Why we fake the mass
Instead of tracking every atom's tug on an electron, we bundle all those forces into one number: pretend the electron is a free particle but heavier or lighter . That trick lets us reuse simple free-particle formulas — and it's exactly what sets the size of N c , N v . See Effective mass .
The star ∗ always means "effective / apparent," never multiplication here.
Two more ingredients hide inside N c and N v ; meet them before Section 9.
Definition Planck's constant
h
h is a fixed constant of nature that sets the "graininess" of quantum states: h ≈ 6.626 × 1 0 − 34 J ⋅ s . In these formulas it appears as h 2 in the denominator, controlling how tightly seats are packed in the band. (The parent note's derivation uses ℏ = h /2 π ; both are the same constant dressed differently.)
Intuition Why a factor of 2 sits in front of
N c , N v
Every spatial energy-state can actually hold two electrons, because an electron carries a spin that can point "up" or "down" — Pauli's rule forbids identical electrons, but opposite spins count as different. So each seat is really a double seat . That extra bookkeeping is the leading 2 in N c = 2 ( … ) and N v = 2 ( … ) . This factor is spin degeneracy ; forget it and you undercount every carrier by half.
N c and N v
N c is the effective number of conduction-band seats per unit volume , as if the whole spread-out band were squeezed into one single shelf sitting right at E c . Likewise N v at E v . Their units are states per unit volume (cm− 3 ) .
N c = 2 ( h 2 2 π m n ∗ k T ) 3/2 , N v = 2 ( h 2 2 π m p ∗ k T ) 3/2
Here the leading 2 is the spin factor (Section 8), h is Planck's constant (Section 8), m ∗ is the effective mass (Section 7), and k T is the thermal energy (Section 4).
Intuition Why bother collapsing the band
The real integral (seats × occupancy over all heights) is messy. In the non-degenerate case (Section 6) it turns out to give the same answer as a much simpler picture: one shelf of N c seats at the edge. So N c is a shortcut number that hides the whole integral. It carries a T 3/2 dependence because warmer crystals effectively offer more reachable seats.
Definition Electron and hole concentrations
n , p
n = electron concentration = the number of mobile electrons in the conduction band per unit volume (units cm− 3 ).
p = hole concentration = the number of empty spots in the valence band per unit volume (units cm− 3 ).
These are the two numbers the entire topic exists to compute.
Now every letter in the master results is earned:
Read the mnemonic on the parent page — "electrons fall from the top, holes rise from the bottom" — and every symbol should now feel like a labelled part, not a mystery.
Band gap Eg = Ec minus Ev
Fermi level EF the water line
Effective states Nc and Nv
Planck constant h and spin factor 2
Non-degenerate Boltzmann tail
Intrinsic ni and np equals ni squared
This map is the whole topic in one glance: energy feeds the bands and the water line; density of states plus occupancy plus effective mass feed the carrier count; the gap and effective states feed n i . Related framing: Intrinsic vs extrinsic semiconductors .
Test yourself — cover the right side and answer out loud.
What does the vertical axis in every band diagram represent? Electron energy E (in eV), height on the ladder.
What are E c and E v ? The conduction band edge (bottom of upper band) and valence band edge (top of lower band).
Define the band gap E g in symbols. E g = E c − E v , the forbidden jump height.
What is special about the Fermi level E F ? It's the 50% occupancy line — states below are mostly filled, above mostly empty.
What are the units and value of k T at 300 K? Energy (eV), with k ≈ 8.62 × 1 0 − 5 eV/K, so k T ≈ 0.0259 eV.
In plain words, what is g ( E ) and its units? Available seats per unit energy and per unit volume; units states⋅ eV− 1 ⋅ cm− 3 .
What does f ( E ) give, and what value does it take at E = E F ? Probability a seat at E is filled; f ( E F ) = 2 1 .
Why does the hole integral use 1 − f ( E ) ? A hole is an empty seat, so its probability is one-minus-occupancy.
What condition lets f ( E ) become the Boltzmann exponential? Non-degenerate case E c − E F ≫ k T , so the "+ 1 " is negligible.
What is h and where does the factor of 2 in N c come from? h is Planck's constant (6.626 × 1 0 − 34 J⋅ s); the 2 is spin degeneracy (two electrons per state).
What does N c represent physically, with units? Effective conduction seats per unit volume (cm− 3 ), as if the band collapsed to one shelf at E c .
What do n and p stand for, with units? Electron and hole concentrations, both in cm− 3 .
Why does n i use e − E g /2 k T (half the gap)? Because
n i = n p and
n p ∝ e − E g / k T , so the square root halves the exponent.