We use the RIOTS4 sample of SMC field OB stars to determine the origin of massive runaways in this low-metallicity galaxy using Gaia proper motions, together with stellar masses obtained from RIOTS4 data. These data allow us to estimate the relative contributions of stars accelerated by the dynamical ejection vs binary supernova mechanisms, since dynamical ejection favors faster, more massive runaways, while SN ejection favors the opposite trend. In addition, we use the frequencies of classical OBe stars, high-mass X-ray binaries, and non-compact binaries to discriminate between the mechanisms. Our results show that the dynamical mechanism dominates by a factor of 2 – 3. This also implies a significant contribution from two-step acceleration that occurs when dynamically ejected binaries are followed by SN kicks. We update our published quantitative results from Gaia DR2 proper motions with new data from DR3.

A fundamental question for theories of massive star formation is whether OB stars can form in isolation. We assess the contribution of any in-situ OB star formation by using 210 field OB stars in the Small Magellanic Cloud (SMC) from the Runaways and Isolated O-Type Star Spectroscopic Survey of the SMC (RIOTS4). We search for tiny, sparse clusters around our target OB stars using cluster-finding algorithms. Employing statistical tests, we compare these observations with random-field data sets. We find that ∼5% of our target fields do show evidence of higher central stellar densities, implying the presence of small clusters. This frequency of small clusters is low and within errors, it is also consistent with the field OB population being composed entirely of runaway and walkaway stars. Assuming this small cluster fraction is real, it implies that some OB stars may form in highly isolated conditions. The low frequency could be caused by these clusters evaporating on a short timescale. However, another interpretation is that the low fraction of small clusters is observed because these form rarely, or not at all, implying a higher cluster lower-mass limit and generally consistent with a relationship between maximum stellar mass (mmax) and the cluster mass (Mcl).