The case for gravitational millilensing in the multiply--imaged quasar B1152+199

Previous Very Long Baseline Interferometry (VLBI) observations of the quasar B1152+199 at 5GHz has revealed two images of a strongly lensed jet with seemingly discordant morphologies. Whereas the jet appears straight in one of the images, the other exhibits slight curvature on milliarcsecond scales. This is unexpected from the lensing solution and has been interpreted as possible evidence for secondary, small-scale lensing (millilensing) by a compact object with a mass of $~10^5$-$10^7\ M_\odot$ located close to the curved image. The probability for such a superposition is extremely low unless the millilens population has very high surface number density. Here, we revisit the case for millilensing in B1152+199 by combining new global-VLBI data at 8.4GHz with two datasets from the European VLBI Network (EVN) at 5GHz (archival) and at 22GHz (new dataset), and the previously published 5GHz Very Long Baseline Array (VLBA) data. We find that the new data with a more circular synthesized beam, exhibits no apparent milliarcsecond-scale curvature in image B. Various observations of the object spanning $\sim$15 years apart enable us to improve the constraints on lens system (thanks also to the improved astrometry resulting from 22GHz observations) to the point that the only plausible explanation left for the apparent curvature is the artifact due to the shape of the synthesized beam.
9 pages, 4 figures, submitted to MNRAS