Paper detail

Comment on "Experimental Determination of the Statistics of Photons Emitted by a Tunnel Junction"

A very recent article [(1) E. Zakka-Bajjani et al., PRL104, 206802 (2010)] has addressed the problem of how the statistics of electrons crossing a quantum conductor influences that of the photons they emit. It is however not clear that the detection used in [1] is sensitive to emitted photons and not to the overall electromagnetic fluctuations, which include vacuum fluctuations. We show that the proof in [1] that non-symmetrized noise is detected is erroneous: supposing that the detection simply takes the square of the voltage, i.e. is sensitive to vacuum fluctuations, leads to identical results. We demonstrate that all the results can be explained in terms of usual gaussian voltage fluctuations instead of photon statistics. Finally, it is found in [1] that the photon noise is gaussian for a tunnel junction, for which the electron counting statistics is poissonian. We show that this result is beyond the experimental sensitivity by at least one order of magnitude, and thus is also incorrect.