The exciton — one electron bound to one hole — is only the beginning. Add one more carrier, and you get a trion. Bind two excitons together, and you get a biexciton. Add more, and the zoo grows: charged biexcitons, exciton-polarons, Fermi polarons, and, in the right conditions, collective phases where many excitons act together. Each species has its own energy, its own optical signature, and its own role in how a material handles light.
The zoo matters because real devices and real experiments are rarely in the pristine single-exciton limit. Materials are doped, gated, or photoexcited. Extra carriers are present. Exciton densities are high enough that interactions cannot be ignored. When you see extra peaks in a spectrum, shoulders that grow with power or gate voltage, or emission that shifts with density, you are often seeing the zoo.
The trion: an exciton plus a spare carrier
A trion is a three-particle bound state: an exciton plus one additional electron or hole. A negatively charged trion consists of two electrons and one hole. A positively charged trion consists of one electron and two holes. In older literature, trions are sometimes called charged excitons, and the terms are used interchangeably.
The trion is bound because the extra carrier is attracted to the opposite charge in the exciton, and the net attraction outweighs the repulsion between the two like charges. The binding energy of the trion — the energy needed to remove the extra carrier and leave a neutral exciton — is much smaller than the exciton binding energy itself, typically a fraction of it. In bulk gallium arsenide, trion binding energies are on the order of a milli-electron-volt or less, visible only at low temperature. In monolayer transition metal dichalcogenides, they are reported to be tens of milli-electron-volts, large enough to be resolved at room temperature in clean samples, though linewidths and disorder can blur them.
Trions are gate-tunable in a way neutral excitons are not. Because they carry net charge, their population depends directly on the free-carrier density. In a field-effect device, sweeping the gate voltage from electron-doped through neutral to hole-doped moves the spectrum from negative trions to neutral excitons to positive trions. That gate dependence is one of the cleanest ways to identify a trion peak: it appears when carriers are present, grows with doping, and fades at charge neutrality.
Trions also have distinct spin and valley structure. In monolayer semiconductors with two valleys, the two electrons in a negative trion can sit in the same valley or in opposite valleys, leading to intravalley and intervalley trions with different energies and optical selection rules. Resolving those fine structures has been a productive subfield, and it connects trion spectroscopy to valley physics, which has its own article in this series.
From a device perspective, trions matter in two ways. First, they are a readout of doping and charge environment — a spectral thermometer for the carrier sea. Second, they are a loss channel for neutral exciton emission: when free carriers are present, neutral excitons can capture a carrier and become trions, shifting emission to lower energy and changing the lifetime. Controlling doping is therefore also controlling which species dominates the light.
The biexciton: two excitons bound together
A biexciton is a bound state of two excitons — two electrons and two holes — analogous to a hydrogen molecule, where two hydrogen atoms bind to form H2. The biexciton binding energy is the energy gained by binding the two excitons together, beyond the energy of two separate excitons. It is typically smaller than the exciton binding energy, again by a substantial factor, and its precise value is sensitive to material parameters, dimensionality, and dielectric environment.
Biexcitons appear at higher excitation density than trions, because forming them requires two excitons to meet. Their emission intensity therefore grows superlinearly with excitation power — roughly quadratically in the simplest picture, compared to roughly linearly for single excitons — which is a practical diagnostic. At low power, the biexciton peak is absent or weak. As power increases, it grows faster than the exciton peak. Power-dependent spectroscopy is thus one of the first tests researchers apply when assigning a new peak.
Biexcitons also have a characteristic emission pattern. When a biexciton recombines, it emits one photon and leaves behind one exciton, which may then emit a second photon. The two photons are correlated in energy and time, which makes biexciton cascades a source of photon pairs. In quantum dots, the biexciton-exciton cascade is a well-studied route to entangled photon pairs, and controlling the fine-structure splitting of the intermediate exciton state is a central engineering challenge for that application.
In extended two-dimensional materials, biexcitons are more delicate than in zero-dimensional dots. Confinement in a dot stabilizes the four-particle complex; in a monolayer, the biexciton must survive scattering and dissociation in a larger, more disordered environment. Clean, encapsulated samples and low temperatures help. Reports of biexciton binding energies in monolayer materials span a range, reflecting differences in material, encapsulation, and assignment, and the literature continues to refine which features are biexcitons, which are trions, and which are defect-related.
Charged biexcitons and larger complexes
Once trions and biexcitons are admitted, the combinatorics continue. A charged biexciton — a biexciton plus an extra carrier, five particles in total — is bound in some materials and has been reported in monolayer semiconductors under appropriate doping and excitation conditions. Larger complexes, sometimes called exciton droplets or multi-exciton complexes, can form at still higher densities, though they quickly merge into the regime where the picture of distinct bound complexes gives way to many-body physics.
The naming can become confusing, and the literature does not always use terms consistently. A useful discipline is to count particles and net charge. A neutral exciton is one electron, one hole, zero net charge. A negative trion is two electrons, one hole, charge minus one. A positive trion is one electron, two holes, charge plus one. A biexciton is two electrons, two holes, zero net charge. A charged biexciton adds one more carrier. When a paper claims a new species, checking that particle count and charge accounting is a good first step.
Exciton-polarons and the Fermi polaron picture
At finite carrier density, a cleaner description than a fixed three-particle trion is sometimes the polaron picture. An exciton immersed in a sea of electrons does not just bind one of them; it is dressed by the sea, shifting its energy and acquiring a cloud of particle-hole excitations. The resulting quasiparticle is called an exciton-polaron, or, in the language borrowed from cold-atom physics, a Fermi polaron.
The polaron picture naturally produces two branches: an attractive branch, lower in energy, that evolves from the trion at low density, and a repulsive branch, higher in energy, that evolves from the neutral exciton. As the carrier density increases, spectral weight transfers from the repulsive branch to the attractive branch, and the energies of both shift. This framework has been particularly productive for interpreting gate-dependent spectra in monolayer semiconductors, where the transition from exciton to trion is not abrupt but continuous, and where the lineshapes and energy shifts carry information about the interaction between the exciton and the Fermi sea.
The choice between trion language and polaron language is not a dispute about what is real. It is a choice of approximation. At very low carrier density, the three-particle trion is a good description. At higher density, the many-body dressing matters, and the polaron picture is more accurate. Good papers state which regime they are in and why the model they use is appropriate.
From complexes to collective phases
At still higher exciton densities, the zoo gives way to collective behavior. Excitons are bosons, to a good approximation when their density is low enough that their constituent fermions do not overlap strongly. Bosons can, in principle, undergo Bose-Einstein condensation — macroscopic occupation of a single quantum state. Exciton condensation has been a long-sought goal, pursued in coupled quantum wells with spatially indirect excitons, in microcavities with polaritons, and in moiré systems.
Closely related is the idea of an excitonic insulator: a phase where excitons form spontaneously, without optical excitation, because the binding energy exceeds the band gap, and the ground state itself is an exciton condensate. The excitonic insulator has been proposed for several materials and remains an active area of investigation, with signatures that are subtle and often contested.
At the other extreme of density, excitons dissolve. When the average spacing between excitons becomes comparable to the exciton size, the bound states overlap, screening increases, and the system crosses over to an electron-hole plasma. That Mott transition is not a sharp line in all experiments, but it sets the boundary beyond which talking about individual excitons stops being useful.
Between those extremes lies a rich middle: exciton liquids, droplets, and correlated states in moiré superlattices, where excitons are trapped in a periodic potential and can form ordered arrays. Moiré excitons have their own article in this series, but they belong to the zoo in the sense that they are what happens when excitons are made to interact in a designed landscape.
How we tell them apart
Assigning a spectral peak to a species is a core skill in exciton spectroscopy, and it relies on several independent checks, not on energy alone.
Gate dependence identifies charged species: trions and charged biexcitons appear with doping and vanish at neutrality. Power dependence identifies multi-exciton species: biexcitons grow superlinearly with excitation. Temperature dependence identifies binding: more weakly bound species disappear first as temperature rises. Polarization and magnetic-field dependence identify spin and valley structure. Time-resolved measurements identify lifetimes: biexcitons and trions often have different decay dynamics than neutral excitons. And comparison across samples — encapsulated versus bare, different substrates, different materials — tests whether a feature tracks the material or the environment.
No single check is decisive. A peak that grows with power could be a biexciton, or it could be a defect state that saturates differently, or a phonon replica. The literature contains reassignments, where a feature first called a biexciton was later shown to be a trion or a defect. That is normal science, and it is why the field values multi-knob experiments that vary gate, power, temperature, and polarization together.
Why the zoo matters beyond spectroscopy
The zoo is not just taxonomy. Each species is a different way to store, move, and emit energy, and each has different device implications.
Trions, because they are charged, can be moved by electric fields in ways neutral excitons cannot. They also couple differently to charge noise and to gates, which matters for electrically tunable emitters and modulators. Biexcitons, because they emit photon pairs, are relevant to quantum light sources. Exciton-polarons, because they encode the interaction between excitons and a Fermi sea, are a probe of many-body physics that can be read optically. Collective phases, if they can be stabilized and controlled, could enable dissipationless transport or novel coherent devices.
For the near term, the most practical lesson is simpler: if you are building an excitonic device, you are building a device that will host the whole zoo, not just the exciton. Doping, excitation density, and temperature will determine which species dominates. Designing for one species while ignoring the others is a common source of surprise.
Sources and further reading
For trions, biexcitons, and many-body complexes, see review articles on excitons in two-dimensional semiconductors in Reviews of Modern Physics, Nature Reviews Materials, and Nature Photonics, which survey trion and biexciton binding energies, gate and power dependence, and the exciton-polaron picture. For biexciton cascades and entangled photon pairs from quantum dots, see reviews on solid-state quantum light sources in Nature Photonics and Reviews of Modern Physics. For exciton condensation, excitonic insulators, and electron-hole plasma physics, see review literature on Bose-Einstein condensation of excitons and polaritons and on dense electron-hole systems. For the Fermi polaron framework applied to semiconductors, see literature on Fermi polarons in ultracold atoms and its adaptation to monolayer semiconductors. Where this article gives ranges or relative magnitudes rather than single values for binding energies, that reflects the spread across materials and measurement conditions reported in that review literature.