In plain words

An that has picked up an extra , which makes it electrically charged. It shows up as a separate, slightly lower-energy glow when a thin carries extra charge.

Going deeper

Left: a neutral exciton, a negative trion with two electrons and one hole, and a positive trion with two holes and one electron, above a photoluminescence spectrum in which the trion peak sits below the exciton peak. Right: the measured binding energies and a caveat about the word. an exciton with one carrier too many X⁰ X⁻ X⁺ neutral two electrons two holes X⁰ X⁻ photon energy add electrons with a gate and the light moves to the lower, charged peak how tightly it is bound monolayer MoS₂: about 20 meV, which is still meaningful at room temperature monolayer MoSe₂: 30 meV, with lines only 5 meV wide below 55 K positive and negative trions cost almost the same, so electrons and holes in these monolayers weigh about the same a caveat on the word once the sheet carries real carriers, the peak is better described as an exciton dressed by a Fermi sea – a polaron – and the two pictures part company at doping
A trion is an exciton that has caught an extra carrier. It emits at lower energy than the neutral exciton by its binding energy, so the balance between the two peaks is a direct read-out of how much charge the monolayer is carrying.

Three particles, one bound state

Two electrons and a hole, or two holes and an electron, can bind together into a single charged complex. In conventional semiconductors such states exist but are so weakly bound that they appear only at low temperature in high-quality . In semiconductors they are robust, for the same reason that neutral excitons are: the carriers are confined to a plane, the is weak because field lines leave the material, and the are heavy.

The trion emits at lower energy than the neutral exciton, because recombination has to leave the extra carrier behind with whatever energy it carries away. The difference between the two peaks is the trion binding energy – around 20 meV in monolayer MoS2, large enough to keep the state relevant at room temperature, and 30 meV in monolayer MoSe2, where the lines are narrow enough – about 5 meV below 55 K – to resolve all three species cleanly.

A gate turns one into the other

Because a trion needs a spare carrier, its brightness tracks the . In a field-effect device the gate does this continuously: at a negative the monolayer holds holes and the positive trion appears, near charge neutrality the neutral exciton dominates, and at a positive gate voltage the negative trion takes over. Watching all three in one sweep is now a standard characterisation of a monolayer device, and it works in reverse as a measurement of where charge neutrality sits.

One detail from those measurements is telling. The charging energies of the positive and negative trions in monolayer MoSe2 come out nearly identical, which implies that electrons and holes in these materials have nearly the same effective mass – a symmetry that is unusual among semiconductors and simplifies a good deal of the physics built on top of it.

When the word stops being right

The three-particle picture is a good description when the extra carriers are few and far apart. Once the monolayer carries a real Fermi sea, the excited state is better thought of as an exciton dressed by all the carriers around it – an attractive Fermi – rather than as a bound complex with one specific partner. The two descriptions agree in the dilute limit, and part company as doping rises: the polaron picture predicts how the peak shifts, broadens and transfers weight with carrier density, which the trion picture does not.

This matters when reading numbers. A quoted trion binding energy depends on the doping at which it was measured and on which model was used to extract it, so values in the literature for the same material vary by more than their error bars. The peak separation is the observable; the binding energy is an interpretation.

For specialists

A charged exciton (an exciton bound to an extra electron or hole), visible as a lower-energy peak in doped monolayers.

Where this comes from

  1. Tightly bound trions in monolayer MoS2 Mak et al. · Nature Materials 12, 207 (2013) cited by 2,894
  2. Electrical control of neutral and charged excitons in a monolayer semiconductor Ross et al. · Nature Communications 4, 1474 (2013) cited by 1,599