Answers are grounded in this paper and cite their sources.
Angle-separated pair photons emitted in spontaneous parametric fluorescence are superimposed at a beamsplitter and detected by coincidence techniques. The coincidence disappears when the photon path lengths to the beamsplitter are equal to within their inverse bandwidth due to fourth-order interference. The photon bandwidth is set by wide apertures placed to satisfy phase-matching conditions at the nonlinear crystal. The resulting broadband photons are short (39-fs full width at half maximum) and show an oscillatory tail resulting from the hard-edge apertures. Two-color photons are created by blocking the central portion of the apertures leading to strong oscillations of the coincidence rate as a function of path-length difference. Both effects are predicted by the theory of fourth-order interference coupled with the detailed phase-matching conditions in the nonlinear crystal used as a source. The two colors overlap at different positions on the beamsplitter. We show how this could be used as a test of local realistic theories by formulating a Bell inequality based on relative phases between the colors.
No takes yet. Share an insight, caveat, or question.
Rarity et al. (1990) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: