Menu Home

Sample Experiments – 2-Photon-Interference by Hong, Ou & Mandel

In 1987, Hong, Ou and Mandel proposed and demonstrated a quantum interference effect that has no classical analog: When two indistinguishable photons impinge on a beam splitter (each on a separate input), they always leave it together at the same output. This is hence called the Hong-Ou-Mandel effect.

To demonstrate this effect, we use the following setup:

Each of the quED outputs is connected to one of the inputs of the quED-HOM. We remove the wave plate from the pump beam of the quED, so that we generate indistinguishable non-entangled photon pairs. Now, we can introduce a temporal delay between the two photons by moving one of the fibre collimators in the quED-HOM. This makes the two photons distiguishable (you can say, one photon arrives at the beam splitter “earlier than the other”). We use this to demonstrate the so-called Hong-Ou-Mandel dip in coincidence counts, by tuning the delay between the two photons:

What happens when the two photons do not hit the beam splitter at the “same time”, so at the left and right edge of the above results? Well the photons do what they would do if the other photon was not there. They can each get reflected or transmitted, independent of what happens to the other one. This means, that in two of four cases, you will see a coincidence event, hence the increase of coincidences in the results when the photons become distinguishable.

So the HOM effect tests for indistinguishability in all possible degrees of freedom. We have built the quED-HOM (and the quED) so that polarization, spatial mode and wavelength overlap should be pretty high. In research, the HOM dip is facilitated to check the indistinguishability of photons, for example consecutive photons being output by a single photon source.

With the motorised version of the quED-HOM, the quCR will plot the coincidence count rates in dependence of the optical path length differences.

What else can be deduced from the width of the dip? Why is there a width anyway? And why are there so many “air quotes” in the text?


What do you need?

  • 1 x quED (high rate version advised)
  • 1 x quED-HOM (for optimal resolution use motorised version)

Connected Experiments