Two floors, and no staircase
Electrons in a solid can't just have any energy they like. The allowed energies come in bands — think of them as floors of a building. The lower floor, the valence band, is where electrons sit when they're bound into the material's bonds: stuck at home. The upper floor, the conduction band, is where electrons are free to roam through the material: free to travel. And there's a strict rule — only electrons that reach the upper floor can carry current.
Between the two floors is a forbidden zone where no electron is allowed to sit. There is no staircase — only a jump. The height of that jump is the band gap, and its size, measured in electron-volts (eV), quietly decides what kind of material you're holding.
The size of the jump defines the material
Everything follows from one number. If the two floors overlap — no gap at all — electrons are always free, so current always flows. That's a conductor. If the jump is small, around an electron-volt, a modest push can send electrons across on demand. That's a semiconductor. If the jump is a cliff — five to nine electron-volts — nothing you can ordinarily do gets an electron over it. That's an insulator. Try each below.
A small hop. Left alone, few electrons cross. Apply a little energy and they jump — leaving a mobile hole behind. That's the switch.
Notice that the conductor and the insulator are, in a sense, the boring cases — both have a fixed answer. The semiconductor is the interesting one, because its answer depends on what you do to it.
Why a small gap means switchable
Here's the number that makes it click. At room temperature, the random thermal jiggle available to an electron is about 0.025 eV. Silicon's gap is 1.1 eV — roughly forty times bigger. So left alone, only a tiny handful of electrons happen to have enough energy to make the jump: pure silicon barely conducts, which is exactly what you want for the "off" state.
But 1.1 eV is not a cliff. Give the material a deliberate push — heat it, shine light on it, or apply a voltage — and you can send a flood of electrons across the gap. Now it conducts: the "on" state. Because the gap is small enough to bridge on command but large enough not to bridge by accident, the material obeys. That controllable jump is the physics beneath the door — a semiconductor you can switch.
Random room-temperature heat carries about 0.025 eV. Silicon's band gap is 1.1 eV — roughly forty times larger. What does that combination give you?
The empty seat: holes
Every jump leaves something behind. When an electron leaps from the valence band to the conduction band, it vacates a spot in the lower floor — an empty seat in the sea of bound electrons. That vacancy is called a hole. And it isn't just an absence: neighbouring electrons shuffle sideways to fill it, so the empty seat appears to drift through the material, behaving like a positive charge carrier moving the other way.
So one jump produces two movers: a free electron above and a mobile hole below. Keeping track of both is the key to the next idea — doping, where engineers deliberately tip the balance to favour spare electrons (N-type) or spare holes (P-type).
Where the gap comes from — and why 1.1 eV is special
The band gap isn't arbitrary; it falls out of how a material's atoms bond. Different materials have different gaps, and that's a feature, not a footnote — it lets engineers pick a material to fit a job. Germanium's gap is 0.66 eV (leaky, temperature-sensitive), silicon's is 1.1 eV, gallium arsenide's is 1.42 eV (great for high frequency and light), and diamond's is 5.5 eV (a superb insulator).
This is band-gap engineering, and you meet it every day. An LED's colour is set by its band gap — the energy of the gap becomes the energy, and therefore the colour, of the light it emits. A solar cell is tuned so sunlight has just enough energy to kick electrons across. And silicon's 1.1 eV sits in the sweet spot for general-purpose switching, which is a big part of why silicon runs the industry.
Diamond's band gap is about 5.5 eV; silicon's is 1.1 eV. Why can't you build ordinary logic switches out of diamond?
Key takeaways
- Electrons live in energy bands: a lower valence band (stuck) and an upper conduction band (free). Only conduction-band electrons carry current.
- The band gap is the energy jump between them, measured in eV. Its size defines the material: overlap → conductor, small gap → semiconductor, huge gap → insulator.
- Silicon's 1.1 eV gap is far bigger than room-temperature heat (0.025 eV) yet small enough for a voltage to bridge — so it switches on command without leaking.
- Each jump leaves a hole — a mobile positive carrier — behind, which doping later exploits.
Frequently asked questions
What is a band gap in simple terms?
Electrons in a solid can only have certain energies, grouped into bands. The lower band (the valence band) holds electrons that are stuck in place, bound to their atoms. The upper band (the conduction band) holds electrons that are free to move and carry current. Between them is a forbidden zone no electron can sit in — the band gap. To carry current, an electron must gain enough energy to jump the gap, from stuck to free. The size of that jump defines whether a material is a conductor, a semiconductor, or an insulator.
How does the band gap decide if a material conducts?
By how hard it is to get electrons into the free (conduction) band. In a conductor the bands overlap — there is no gap — so electrons are always free and current always flows. In an insulator the gap is huge (roughly 5–9 eV), a cliff no ordinary energy can drive an electron over, so nothing conducts. In a semiconductor the gap is small (around 1 eV), a hop — so a modest push of heat, light, or voltage sends electrons across on command. That is why only semiconductors can be switched.
What is the band gap measured in, and what is an electron-volt?
Band gaps are measured in electron-volts (eV). One electron-volt is the tiny amount of energy an electron gains moving through a one-volt difference — a convenient unit at the atomic scale. For reference, random room-temperature thermal energy is about 0.025 eV, silicon's band gap is about 1.1 eV, and an insulator like diamond is about 5.5 eV. The ratio between the everyday thermal jiggle and the gap is what tells you whether electrons cross easily, rarely, or essentially never.
What is silicon's band gap and why does its value matter?
Silicon's band gap is about 1.1 eV, and that value is a Goldilocks number. It's large enough that random room-temperature heat (about 0.025 eV) rarely pushes electrons across by accident, so the material doesn't leak when it should be off. Yet it's small enough that a modest applied voltage easily drives electrons across, so the material switches on cheaply and quickly. That balance between 'won't leak' and 'switches easily' is a big part of why silicon runs the industry.
What is a hole in a semiconductor?
When an electron jumps from the valence band up to the conduction band, it leaves an empty spot behind — a vacancy in the sea of bound electrons. That vacancy is called a hole. Neighbouring electrons shuffle into it, which makes the hole appear to drift in the opposite direction, behaving like a positive charge carrier. So one jump creates two carriers: a free electron above and a mobile hole below. Doping deliberately tips the balance toward one or the other.
See how you’d score in a real interview.
These deep dives build the foundation. The readiness check goes further: questions drawn from real semiconductor interviews, scored instantly, with a clear read on where you stand and what to work on next.
Take the free readiness checkFree, from semiconductor.cv — the career platform behind this series.
Preparing for semiconductor interviews? Take the free readiness check →