183. Constraining photon-hidden photon oscillations via CMB

This is a guest blog post by Alessandro Mirizzi, Research Fellow at Max Planck Institute of Physics, Munich. Dmitry.

Recently, Javier Redondo, Guenter Sigl and I have completed a study on the possibility to constrain photon-hidden photon oscillations using the extremely precise measurements of the cosmic microwave backgroung (CMB). Here I discuss our results, presented in arXiv:0901.0014.

A hidden photon is the gauge boson of a gauge hidden symmetry under which all the SM fields are uncharged, and thus remains hidden in our world. At energies above the electroweak scale, particles charged under the hidden and the SM hypercharge are likely to exist and act as messengers between the two “sectors” of low-energy physics. Their effects will produce effective operators in the low-energy theory that contains the SM and the hidden photon. Typically, these operators are irrelevant, suppressed by the heavy particle masses. But there is still one marginal operator, the so called kinetic mixing whose effects are not in principle suppressed by heavy masses and thus can lead to relatively large effects. This mixing leads to oscillations between photons and hidden photons, analogous to the observed oscillations between neutrinos of different flavors.

Recently, new intriguing ideas and experimental techniques have been proposed to achieve a possible detection of these elusive particles. For example, hidden photons could be produced through kinetic mixing with solar photons. In this case, it would be possible to search for an hypothetical hidden photon flux from the Sun with Super-Kamiokande and with the CAST axion detector. The expected signature of hidden photons in Super-Kamiokande water Cherenkov detector is particularly intriguing. Since hidden photons, coming from the Sun, are long-lived noninteracting particles, they would penetrate the Earth shielding and oscillate into real photons in the vacuum space between the photomultiplier (PMT) envelope and the photocathode in Super-Kamiokande. The photon then would convert in the photocathode into a single photoelectron which would be detected by the PMT. This is not possible for hidden photons coming from the water tank, since the presence of the medium suppresses the hidden photon-photon oscillations. The result is an increase of the counting rate of those PMTs that are “illuminated” by the Sun from the back, in comparison to those facing the Sun.

Possible signatures of hidden photons could also be achieved analyzing the existing high-energy gamma-ray data taken with several Cherenkov telescopes. Also laser experiments, like ALPs, GammeV, BMV searching for light shining through walls, could discover hidden photons.

In 1983, Georgi, Ginsparg and Glashow, in Nature 306, 765 – 766 (22 December 1983), proposed the existence of a hidden photon mixing with the ordinary photons to explain the discrepancy between theoretical and observed CMB spectrum. After that time, measurements of CMB have reached an astonishing accuracy and no tension between theory and experiments remains. In particular, the Far Infrared Absolute Spectrophotometer (FIRAS) on board of the Cosmic Background Explorer (COBE) confirmed the blackbody nature of the CMB spectrum at better than 1 part in . Therefore, CMB now is one of the most powerful probes used to constrain exotic physics. Any new particle beyond the standard model which can interact with photons can potentially distort the blackbody spectrum, providing us with a glimpse of the existence of such a particle. In this respect, the CMB provides an excellent probe for photon oscillations into low mass hidden photons, since the baseline is the longest at our disposal, namely the whole universe.

The purpose of our paper has been to revisit the CMB bounds on photons-hidden photons mixing using the most precise available observations of the CMB spectrum. The already existing constraints did not properly take into account the refractive properties of the primordial plasma in which photons propagate and were thus incomplete. These medium effects are especially important when we realize that a resonant photon-hidden photon conversion is possible, similar to the Mikheev-Smirnov-Wolfenstein effect in the neutrino case. This resonant conversion is much stronger than the vacuum oscillations considered so far and, therefore, the bounds we obtain are much stronger.

Requiring the distortions of the CMB induced by the photon-hidden photon mixing to be smaller than experimental upper limits, this leads to a bound on the mixing angle for hidden photon masses between eV and eV. This low-mass and low-mixing region of the hidden photon parameter space was previously unconstrained. As a result of our new bound, it is unlikely that hidden photons with masses smaller than eV can play a cosmological role. Conversely, for meV masses, resonant photon-hidden photon oscillations happen after nucleosynthesis but before CMB decoupling, increasing the effective number of neutrinos but also the baryon to photon ratio with interesting cosmological consequences. The mixing angles required for this effect could be probed in current laboratory experiments: Fiat (hidden) lux!