How to make a pure downconversion photon

• Post by: admin
• January 7, 2018
• No Comment

It’s been almost a year since I last added some news here but that doesn’t mean nothing has happened in the lab. We’ve made some very good progress on one of our major goals, the creation of spectrally pure single photons via parametric downconversion.

This progress is summarised in one theory paper published earlier last year and a more recent preprint describing our experimental results. The short summary is that we found not one but two new algorithms for creating downconversion photons with joint Gaussian spectra — and both outperform all previous results in particular for short nonlinear crystals. Although often overlooked, nonlinearity tailoring is the third and equally crucial requirement of producing spectrally pure photons via what is inaccurately summarised as group-velocity matching, the other two being the actual group-velocity matching and the matching of the downconversion and pump bandwidth.

The new domain engineering algorithm was put to the test experimentally and the results look pretty good. Without any spectral filtering, we measure a lower bound on spectral purity close to 91% which is pretty close to the ideal target purity of just above 95% (which could be higher but isn’t due to known and fixable issues with the pump laser). For comparison, a standard crystal with a rectangular nonlinear profile restricts the purity to under 81%. This means we can ditch the frequency filters from our experiments and ultimately reach much higher multi-photon brightness — the key to at least in principle scalable photonic quantum technology.