Wiki / Sources / LED

LED

Emits a collimated beam of incoherent light with a single-colour or two-band white LED spectrum, for illumination, filtering, and colour separation.

Open in the canvas →

A live trace, not a picture of one — but this preview is not interactive. Open it in the canvas to move things, change parameters, and save or export your own version.

In the real world

A light-emitting diode is a semiconductor junction that emits when electrons and holes recombine across its band gap. The photon energy follows the gap, so the material sets the colour: InGaN for blue and green, AlInGaP for amber and red, AlGaAs and GaAs towards the near infrared. The emission is spontaneous, not stimulated, which gives an LED its two defining properties. Its spectrum is a smooth band tens of nanometres wide, against the fraction of a nanometre of a laser diode, and its light is incoherent: each recombination is independent of the next.

A white LED is not a white emitter. It is a blue die coated with a phosphor, usually cerium-doped YAG, that absorbs part of the blue and re-emits a broad yellow band. The spectrum therefore has two parts — a narrow blue peak left over from the die and a broad phosphor band — and the balance between them is what makes one white LED look cool and another warm.

The die is an extended emitter radiating into a full hemisphere, so an LED meant for a bench is sold behind its own optics: a moulded dome, a reflector, or an aspheric condenser. That collimator turns the hemisphere into a beam, but never a perfectly parallel one, because the die has a size. The residual divergence is roughly the die size divided by the collimator's focal length.

λ≈hcEg\lambda \approx \frac{hc}{E_g}
The emission wavelength follows the band gap of the semiconductor.
θ≈df\theta \approx \frac{d}{f}
Residual full divergence of a collimated LED: die size d over collimator focal length f. A 1 mm die behind a 20 mm lens leaves about 50 mrad, close to 3°.

In OpticalSetup

The LED is a packaged source: the die and its collimator are one element, and what leaves the housing is a collimated beam of the width you set, made of ordinary rays that travel until something stops them. This is what separates it from the Point source, which is the bare emitter — isotropic, and only useful once a lens or a mirror collects it.

Choose a preset or a custom centre wavelength and width. Each single-colour preset is one Gaussian band. The white preset is two: a blue peak near 450 nm and a phosphor band near 580 nm, carried as one spectrum with both parts in it. Filters, dichroics, etalons and the spectrometer integrate against that shape, so a long-pass filter at 500 nm removes the blue peak and leaves the phosphor band, and a prism or grating fans the whole spectrum out by colour. The undispersed white beam is drawn in the pale mixed-light colour used for any broadband beam.

The LED is an incoherent source. It has no coherence setting, and its beams never interfere: where two of them meet on a detector, their powers add. An interferometer built with an LED shows no fringes.

Simplified vs. reality

The presets are illustrative shapes, not any manufacturer's datasheet, and the spectrum does not shift with drive current or temperature. The beam leaves perfectly collimated: the residual divergence a real die size imposes is not modelled, and neither is the non-uniform, die-shaped intensity across a real LED beam. The emission is steady; pulsed or modulated drive is not represented. The light is unpolarized.

Related components

Further reading