How it works
An LED is a forward-biased PN junction made from a direct bandgap semiconductor. When electrons in the conduction band recombine with holes in the valence band, they release energy as a photon rather than heat — unlike silicon, which is an indirect bandgap material and releases energy as phonons (heat). Photon wavelength λ = hc/E_g; for GaP (green, E_g ≈ 2.2 eV), λ ≈ 565 nm; for InGaN (blue/white), E_g is tunable between 2.7 and 3.4 eV. White LEDs use a blue InGaN die coated with yellow phosphor (Ce-doped YAG) to produce broadband white emission.
Key points to remember
Forward voltage V_F varies by material: ~1.8–2.2 V for red/yellow (GaAsP/AlGaInP), ~3.0–3.5 V for blue/white (InGaN), ~1.2–1.5 V for infrared (GaAs). Luminous intensity in mcd is not proportional to forward current beyond the recommended operating point — LED efficiency (lm/W) actually decreases at high drive currents due to efficiency droop. Maximum forward current for standard 5 mm LEDs is typically 20–30 mA; absolute maximum at 50 mA causes permanent degradation. Peak emission wavelength shifts slightly with temperature and drive current. Direct bandgap is mandatory for LEDs because indirect bandgap materials have very low radiative recombination probability.
Exam tip
The examiner always asks you to explain why silicon cannot be used to make LEDs — the answer is that silicon has an indirect bandgap, so conduction band electrons cannot recombine radiatively with valence band holes without also changing crystal momentum via a phonon interaction.