Bridging the gap between stars and planets
The formation mechanisms of brown dwarfs (BDs) and free-floating planets (FFPs) are poorly constrained. The two main competing scenarios include the star-like scenario, where these objects correspond to the least massive gravitationally unstable fragments of molecular clouds, and the planet-like scenario, where these objects form in a circumstellar disk and are subsequently ejected after a dynamical interaction. A new generation millimeter interferometer, such as ALMA2040, is required to disentangle the relevance of these scenarios. For this, ALMA2040 must provide the appropriate performance to explore BD/FFP formation from the cloud scale (~10 pc) down to the disk scale (~10 au). At cloud scales, ALMA2040 must detect a ~1 Mjup embedded FFP in the outer Galaxy (~10 kpc) in 1 h, or the same object in nearby clouds (<~ 400 pc) in about tens of seconds so that large-scale surveys of entire clouds can be performed. This would provide key censuses to compare the results with the predictions from each scenario. At disk scales, ALMA2040 must detect CO outflows from first hydrostatic cores in the substellar regime and potential molecular outflows from ~1 Mjup FFPs in nearby clouds, both with specific predictions in the star-like scenario. In addition, the planet-like scenario predicts a disk mass-to-star mass ratio for BDs larger than the star-like case, and testing this requires detecting disks of ~2 Earth masses around young BDs. Furthermore, the predicted emission for the star-like case of complex organic molecules in proto-BDs at 400 pc must be detected with ALMA2040, requiring a factor of 10 increase in line sensitivity at a spatial resolution of 5-10 au. Finally, ALMA2040 must detect Earth-like protoplanets around young BDs at 400 pc, in order to provide key observational constraints on the radial drift behaviour and planet formation mechanisms in the substellar regime.