Star formation and quenching among the most massive galaxies at z ~ 1.7

We have conducted a detailed object-by-object study of a mass-complete (M ≥ 10M) sample of 56 galaxies at 1.4 ≤ z ≤ 2 in the Great Observatories Origins Deep Survey-South field, showing that an accurate deblending in 24 μm images is essential to properly assign to each galaxy its own star formation rate (SFR), whereas an automatic procedure often fails. This applies especially to galaxies with SFRs below the main sequence (MS) value, which may be in their quenching phase. After that, the sample splits evenly between galaxies forming stars within a factor of 4 of the MS rate (~45 per cent), and sub-MS galaxies with SFRs ~10-1000 times smaller (~55 per cent). We did not find a well-defined class of intermediate, transient objects below the MS, suggesting that the conversion of a massive MS galaxy into a quenched remnant may take a relatively short time (< 1 Gyr), though a larger sample should be analysed in the same way to set precise limits on the quenching time-scale. X-ray detected AGNs represent a ~30 per cent fraction of the sample, and are found among both star-forming and quenched galaxies. The morphological analysis revealed that ~50 per cent of our massive objects are bulge-dominated, and almost allMS galaxies with a relevant bulge component host an AGN. We also found sub-MS SFRs in many bulge-dominated systems, providing support to the notion that bulge growth, AGN activity and quenching of star formation are closely related to each other.