How many quantum dots are in a volume smaller than the wavelength of light? You can't image them individually — optical microscopy can't resolve objects separated by less than about half a wavelength. Electron microscopy can resolve them but destroys the sample. Fluorescence intensity scales with number, but calibration requires knowing the brightness of a single dot, which varies from dot to dot. Counting emitters in a subwavelength volume is a measurement problem that conventional optics cannot solve.
Ni, Yu, Li, and collaborators (arXiv 2602.22677, February 2026) solve it using photon correlations. They confine CdSe/CdS/ZnS colloidal quantum dots inside polystyrene capsules smaller than the emission wavelength. The dots emit non-polarized, homogeneously broadened light — they are effectively identical quantum emitters in the Dicke sense. Their collective emission follows Dicke's superradiance theory: N identical emitters in a subwavelength volume radiate collectively, and the photon statistics of the collective emission encode the number of emitters.
The key observable is the second-order photon correlation function g²(0) — the probability of detecting two photons simultaneously, normalized by the probability expected for uncorrelated light. A single quantum emitter has g²(0) = 0: it can't emit two photons at once (antibunching). Two emitters have g²(0) = 0.5. Three have g²(0) = 0.67. The value approaches 1 from below as N increases. Each integer N produces a distinct g²(0), so measuring g²(0) counts the emitters.
The collective lifetime adds a second observable. Dicke superradiance accelerates the spontaneous emission: N emitters decay N times faster than a single emitter. The lifetime shortening is proportional to 1/N. Measuring the decay time gives a second, independent estimate of N. The two observables — photon statistics and collective lifetime — constrain the emitter number together, providing redundancy and improving confidence.
The method works for 1 to 10 quantum dots per capsule, which is the range relevant for single-molecule labeling, nanoscale sensing, and quantum technology applications. Beyond about 10, the g²(0) values become too close together to resolve (they all approach 1), and other methods are needed.
The physics exploits a property that's usually a nuisance: the indistinguishability of the emitters. If the quantum dots had different emission wavelengths, different polarizations, or different positions resolvable by the detector, their photon correlations wouldn't follow the Dicke model. The method works precisely because the emitters are too close together and too similar to tell apart by any other means. The indistinguishability that prevents individual identification enables collective counting.