Abstract

A two-photon correlation experiment of the kind that was proposed by Franson and has recently been carried out by two groups is analyzed. In this experiment two simultaneously produced photons are directed along two paths to two photodetectors without mixing. An unbalanced interferometer with delay time that is much longer than the coherence time of the light is inserted in each path. Yet the rate of two-photon detection by the two detectors in coincidence exhibits interference as the path difference is changed. The question of whether classical field theory can account for these observations is discussed. We show that no ergodic classical field is able to reproduce the experimental results, although a nonergodic field can give rise to interference through the artifice of generating an ensemble by initializing repeatedly.

© 1990 Optical Society of America