2008MNRAS.384..719H

The first images of 6.7-GHz methanol masers in the massive star-forming regions DR21(OH) and DR21(OH)N are presented. By measuring the shapes, radial velocities and polarization properties of these masers, it is possible to map out the structure, kinematics and magnetic fields in the molecular gas that surrounds newly formed massive stars. The intrinsic angular resolution of the observations was 43 mas (∼100 au at the distance of DR21), but structures far smaller than this were revealed by employing a non-standard mapping technique. By plotting the positions of the Gaussian-fitted maser emission centroids in each velocity channel, the internal velocity gradients of the masers were investigated at very high spectral and spatial resolution. This technique was used in an attempt to identify the physical structure (e.g. disc, outflow, shock) associated with the methanol masers. Two distinct star-forming centres were identified. In DR21(OH), the masers had a linear morphology, and the individual maser spots each displayed an internal velocity gradient in the same direction as the large-scale structure. They were detected at the same position as the OH 1.7-GHz ground-state masers, close to the centre of an outflow traced by CO and Class I methanol masers. The shape and velocity gradients of the masers suggest that they probably delineate a shock. In DR21(OH)N, the methanol masers trace an arc with a double-peaked profile and a complex velocity gradient. This velocity gradient closely resembles that of a Keplerian disc. The masers in the arc are 4.5 per cent linearly polarized, with a polarization angle that indicates that the magnetic field direction is roughly perpendicular to the large-scale magnetic field in the region (indicated by lower angular resolution measurements of the CO and dust polarization). The origin and nature of these maser structures are considered within the context of what is already known about the region. The suitability of channel-by-channel centroid mapping is discussed as an improved and viable means to maximize the information gained from the data.