Context. Diffuse radio emission in galaxy clusters, namely radio halos and radio relics, is usually associated with merger events. Despite the tremendous advances in observations in the last decades, the particle (re-)acceleration and magnetic field amplification mechanisms and the connection with the stage and geometry of the cluster merger are still uncertain.
Aims: In this paper, we present the peculiar case of PSZ2 G091.83+26.11 at z = 0.822. This cluster hosts a megaparsec-scale radio halo and an elongated radio source whose morphology resembles that of a radio relic. However, the location of this diffuse radio source with respect to the intracluster medium (ICM) distribution and to the cluster centre is not consistent with a simple merger scenario.
Methods: We use Karl Jansky Very Large Array data at 1-4 GHz to investigate the spectral and polarisation properties of the diffuse radio emission. We combine these data with previously published data from the Low Frequency Array (LOFAR) in the 120-168 MHz band and from the upgraded Giant Metrewave Radio Telescope (uGMRT) at 250-500 and 550-900 MHz. Finally, we complement the radio data with Chandra X-ray observations in order to compare the thermal and non-thermal emission of the cluster.
Results: The elongated radio emission east of the cluster is visible up to 3.0 GHz and has an integrated spectral index of α144 MHz3.0 GHz = −1.24 ± 0.03, with a steepening from −0.89 ± 0.03 to −1.39 ± 0.03. These values correspond to Mach numbers ℳradio, int = 3.0 ± 0.19 and ℳradio, inj = 2.48 ± 0.15. Chandra data revealed a surface brightness discontinuity at the location of the radio source with a compression factor of C = 2.22 −0.30+0.39 (i.e. MXray = 1.93−0.32+0.42). We also found that the source is polarised at GHz frequencies. Using QU-fitting, we estimate an intrinsic polarisation fraction of p0 ∼ 0.2, a Rotation Measure of ∼50 rad m−2 (including the Galactic contribution), and an external depolarisation of σRM ∼ 60 rad m−2. The polarisation B-vectors are aligned with the major axis of the source, suggesting magnetic field compression. Hence, we classify this source as a radio relic. Finally, we found a trend consistent with a linear or super-linear correlation between the non-thermal and thermal emission.
Conclusions: We propose an off-axis merger and/or multiple merger events to explain the position and orientation of the relic with the respect to the ICM emission. Given the properties of the radio relic, we speculate that PSZ2 G091.83+26.11 is in a fairly young merger state.
