
The probability that a detected event corresponds to a true single photon, not multiple.In quantum optics, purity is a measure of how confidently a detected signal corresponds to a single photon rather than multiple photons arriving together. When a detector clicks, it could be because one photon arrived, or because two or more photons arrived simultaneously and triggered the same response. Purity quantifies the probability that the click was caused by exactly one photon. This matters because many quantum technologies — from secure communication to quantum computing — rely on single photons as information carriers. If a source emits multiple photons when you wanted one, the security or fidelity of the protocol degrades.
The concept of photon antibunching and the g²(0) criterion were introduced in the 1970s, with the first experimental demonstration of antibunching by H. Jeff Kimble, Mario Dagenais, and Leonard Mandel in 1977 using sodium atoms.
Purity is typically measured using correlation experiments, such as the Hanbury Brown–Twiss setup, where a beam is split and sent to two detectors. By measuring the rate of coincidences (both detectors clicking at the same time) versus single detections, one can extract the second-order correlation function g²(0). For an ideal single-photon source, g²(0) = 0, meaning no coincidences. In practice, a source is considered good if g²(0) is well below 0.5, which indicates that the probability of multi-photon events is low. The purity is then related to this value: the closer g²(0) is to zero, the higher the purity.
Purity is the key quality metric for single-photon sources used in quantum key distribution, quantum random number generation, and linear optical quantum computing.
A perfect single-photon source would have g²(0) = 0, but even the best current sources achieve values around 0.0001, meaning that out of ten thousand emissions, only one contains two photons.
















