New constraints on very light pseudoscalars

There is quite a number of puzzling phenomena in astrophysics that could be clarified with the existence of very light scalar or pseudoscalar particles, commonly refered to as axion-like particles (a.k.a. ALPs). One of the main reasons is that these hypothetical particles can have a coupling to photons inside external magnetic fields, as in the axion case [1], and can lead to observable signatures in every measurable property of light: its intensity, its spectrum, or its polarisation. In this work, we want to present polarimetry as a tool to search for (or, at least, to constrain) the existence of such nearly massless particles. To illustrate this, we use the sample of 355 high-quality polarisation measurements which confirmed the existence of large-scale correlations of (linear) polarisation position angles of visible light from quasars in some regions of the sky [2]. It was thought that these largescale alignments could be naturally explained by ALP-photon mixing, as their phenomenology predicts the creation of linear (plin) and circular (pcirc) polarisations. However, the recent observation of vanishing circular polarisation [3] is a problem for this hypothesis; see [4] and references therein. As ALPs cannot explain the alignments, these data can be exploited to constraint axion-likeparticle parameters. We are going to consider objects located behind the Virgo supercluster, as we have information on plin and pcirc in this region. One can then consider ALP-photon mixing in the magnetic field of that supercluster, and check for which parameters the generated polarisation would match (or in this case contradict) the observations. We use the following: