The puzzling properties of the magnetic O star Tr16-22

Abstract

Context. The detection of bright, hard, and variable X-ray emission in Tr16-22 prompted spectropolarimetric observations of this star, which in turn led to the discovery of a surface magnetic field. Aims. We want to further constrain the properties of this star, in particular to verify whether X-ray variations are correlated to changes in optical emission lines and magnetic field strength, as expected from the oblique rotator model that is widely accepted for magnetic O stars. Methods. We have obtained new low-resolution spectropolarimetric and long-term high-resolution spectroscopic monitoring of Tr16-22, and we also analyse new, serendipitous X-ray data. Results. The new X-ray observations are consistent with previous data, but their addition does not help to solve the ambiguity in the variation timescale because of numerous aliases. No obvious periodicity or any large variations are detected in the spectropolarimetric data of Tr16-22 obtained over three months. The derived field values appear to be in line with previous measurements, suggesting constancy of the field (though the possibility of small, short-term field variations cannot be excluded). Variations in the equivalent widths of Hα are very small, and they do not appear to be related to the X-ray timescale; the overall lack of large variations in optical emission lines is consistent with the magnetic field constancy. In addition, variations of the radial velocities indicate that Tr16-22 is probably a SB1 binary with a very long period. Conclusions. Our new measurements of optical emission lines and magnetic field strength do not show an obvious correlation with X-ray variations. Our current data thus cannot be interpreted in terms of the common model, which assumes the electromagnetic emission associated with a wind confined by a dipolar field tilted with respect to the rotation axis. However, the sampling is imperfect and new data are needed to further constrain the actual periodicity of the various observed phenomena. If inconsistencies are confirmed, then we will need to consider alternative scenarios.