English
Menu

Direct vs Indirect Bandgap: Key Differences Explained

In a direct-bandgap semiconductor, the bottom of the conduction band and the top of the valence band sit at the same position in momentum (wave-vector) space, so an electron and hole can recombine and emit light directly. In an indirect-bandgap semiconductor the two positions are offset, so emitting or absorbing light requires a phonon (lattice vibration) to take part at the same time and make up the momentum. That is why direct-bandgap materials such as GaAs, InP and GaN are used in LEDs and semiconductor lasers, while the indirect-bandgap Si and Ge are poor light emitters but are widely used in photodetectors and LSIs.

AspectDirect bandgapIndirect bandgap
Band structureThe conduction-band bottom and valence-band top are at the same position in momentum spaceThe conduction-band bottom and valence-band top are offset in momentum space
Light emission/absorption and phononsTransitions can occur without phononsA phonon must take part at the same time to make up the momentum
Emission efficiencyHigh. Radiative recombination is likelyExtremely low. Most recombination is non-radiative and becomes heat
Minority-carrier lifetime (typical)Short, on the order of nsLong. In high-quality silicon it reaches µs to ms
Absorption near the band edgeRises steeply from the absorption edge (about 10⁴ cm⁻¹ or more)Weak near the absorption edge, growing gradually as photon energy increases
Absorption depthShallow. In GaAs, a thickness of about 1 to 2 µm absorbs most of sunlightDeep. In silicon, light of about 1 µm wavelength penetrates 100 µm or more
In photodetectorsA thin absorption layer is enough, suiting high-speed detectorsNeeds a thick absorption layer or a light-trapping structure
Typical materialsGaAs (about 1.42 eV), InP (about 1.34 eV), GaN (about 3.4 eV), alloys such as InGaNSi (about 1.12 eV), Ge (about 0.66 eV), SiC, GaP
Main usesLEDs, semiconductor lasers, high-efficiency solar cells, photodetectors for communicationsLSIs, power devices, photodiodes, image sensors, solar cells

Figures are typical orders of magnitude and vary by process generation and product.

When to use which

Direct bandgap
When converting electricity into light. Used in the emitting layers of LEDs and semiconductor lasers, and in photodetectors and high-efficiency solar cells that need strong absorption in a thin layer.
Indirect bandgap
When light emission is not needed. Besides LSIs and power devices, silicon is used for light detection in photodiodes, image sensors and solar cells.
Combined
In silicon photonics, waveguides, modulators and Ge photodetectors are built on silicon, while the light source is a direct-bandgap laser, such as an InP-based one, bonded onto the silicon or combined as a separate chip.

FAQ

How can indirect-bandgap silicon absorb light?

In absorption, a transition can occur if a phonon is absorbed or emitted to make up the momentum. Because the probability is low, absorption is relatively weak, but with enough thickness enough light is absorbed. This is why silicon photodiodes and image sensors detect visible light well.

How do electrons and holes recombine in an indirect-bandgap semiconductor?

Radiative recombination, which emits light, is unlikely, so most recombination goes through crystal defects or impurity levels, such as SRH recombination, and the energy is lost as heat rather than light. The long minority-carrier lifetime in silicon is also due to radiative recombination being rare.

Why is Ge used in photodetectors for communications even though it has an indirect bandgap?

Ge has an indirect bandgap, but its direct gap of about 0.8 eV is close to its indirect gap (about 0.66 eV). It therefore strongly absorbs light at communication wavelengths up to around 1.55 µm, and because it can be grown readily on silicon, it is used in silicon photonics photodetectors.

What determines the emission wavelength?

Mainly the bandgap of the emitting layer: the wavelength λ (nm) is roughly 1240 ÷ Eg (eV). GaAs, at about 1.42 eV, emits near-infrared light at about 870 nm, while GaN-based materials emit blue to ultraviolet light. Changing the alloy composition tunes the bandgap; in InGaN, adding more indium makes the wavelength longer.

Check your understanding

What type of band structure does gallium arsenide (GaAs) have?

  1. ADirect band gap
  2. BIndirect band gap
  3. CNo band gap
Show answer and explanation

Answer: A. Direct band gap

GaAs is a direct-gap semiconductor: its conduction-band minimum and valence-band maximum coincide in momentum space, so electrons and holes recombine radiatively without phonons and emit light efficiently. It is used in infrared LEDs and lasers (optical discs, communications, VCSELs for face authentication). Silicon's indirect gap requires phonons, making emission very inefficient. GaAs also absorbs strongly, so thin layers suffice in high-efficiency solar cells.

What type of band structure does silicon have?

  1. ADirect band gap
  2. BIndirect band gap
  3. CMetallic, with no band gap
Show answer and explanation

Answer: B. Indirect band gap

Silicon is an indirect-gap semiconductor: the conduction-band minimum and valence-band maximum sit at different positions in momentum (wave-vector) space. For an electron and hole to recombine and emit light, a phonon (lattice vibration) must take part to supply the momentum, so light emission is very inefficient. LEDs and lasers therefore use direct-gap materials such as GaAs, InP and GaN. Absorption, however, can occur with phonon assistance, so silicon works well as a photodetector.

In an indirect-gap semiconductor such as silicon, why is the probability of electron-hole recombination with light emission low?

  1. ABecause the conduction band minimum and valence band maximum have different momenta, so the transition requires a phonon
  2. BBecause the hole mobility is lower than the electron mobility
  3. CBecause the band gap is too large
Show answer and explanation

Answer: A. Because the conduction band minimum and valence band maximum have different momenta, so the transition requires a phonon

In an indirect-gap semiconductor, the conduction-band minimum and valence-band maximum are separated in momentum (wave-vector) space. Photons carry almost no momentum, so a phonon must take part in the transition, making radiative recombination orders of magnitude less likely than in direct-gap materials. Carriers instead recombine non-radiatively, e.g. via SRH, releasing heat, so silicon is a poor light emitter. This is why silicon photonics bonds materials such as InP for the light source.

  1. What mainly determines the emission color (wavelength) of an LED?
  2. In silicon photonics, why are compound semiconductors combined with silicon for the laser light source?
  3. Which is a correct reason germanium (Ge) is widely used for telecom-wavelength photodetectors in silicon photonic circuits?

Practice Semiconductor physics questions

Key terms explained

Direct band gap
A band structure in which the bottom of the conduction band and the top of the valence band are at the same wave number. Electron-hole recombination emits light efficiently. Example: GaAs.
Indirect band gap
A band structure in which the bottom of the conduction band and the top of the valence band occur at different momenta. Transitions need a phonon, so light emission is inefficient. Silicon and germanium are typical examples.
Band gap
The energy difference between the top of the valence band and the bottom of the conduction band. About 1.12 eV for silicon at room temperature. A larger band gap means better tolerance to high temperatures and voltages.
GaAs (gallium arsenide)
A III-V compound semiconductor of gallium and arsenic with a direct band gap of about 1.42 eV. Its high electron mobility makes it useful for high-frequency transistors and infrared LEDs and lasers.
Photodiode
A diode that generates charge (current) when it receives light. Used as the light-receiving element in image sensor pixels.

Related comparisons