Context. This paper is the third in a series of NH3 multilevel imaging studies in well-known, high-mass star-forming regions. The main goal is to characterize kinematics and physical conditions of (hot and dense) circumstellar molecular gas around O-type young stars. Aims: We want to map at subarcsecond resolution highly excited inversion lines of NH3 in the high-mass star-forming region W51 Main (distance = 5.4 kpc), which is an ideal target to constrain theoretical models of high-mass star formation. Methods: Using the Karl Jansky Very Large Array (JVLA), we mapped the hot and dense molecular gas in W51 Main with ~0.2 arcsec-0.3 arcsec angular resolution in five metastable (J = K) inversion transitions of ammonia (NH3): (J,K) = (6, 6), (7, 7), (9, 9), (10, 10), and (13, 13). These lines arise from energy levels between ~400 K and ~1700 K above the ground state. We also made maps of the (free-free) continuum emission at frequencies between 25 and 36 GHz. Results: We have identified and characterized two main centers of high-mass star formation in W51 Main, which excite hot cores and host one or multiple high-mass young stellar objects (YSOs) at their centers: the W51e2 complex and the W51e8 core (~6%26#39;%26#39; southward of W51e2). The former breaks down into three further subcores: W51e2-W, which surrounds the well-known hypercompact (HC) HII region, where hot NH3 is observed in absorption, and two additional dusty cores, W51e2-E (~0.8 arcsec to the East) and W51e2-NW (~1%26#39;%26#39; to the North), where hot NH3 is observed in emission. The velocity maps toward the HC HII region show a clear velocity gradient along the east-west in all lines. The gradient may indicate rotation, although any Keplerian motion must be on smaller scales ( is observed in emission. The velocity maps toward the HC HII region show a clear velocity gradient along the east-west in all lines. The gradient may indicate rotation, although any Keplerian motion must be on smaller scales (?, assuming [NH3]/[H2] = 10-7) with respect to the mass of the central YSO estimated from radio luminosity (%26gt;20 M?), both indicate that the central YSO has already accreted most of its final mass. On the other hand, the nearby W51e2-E, while not showing evidence of rotation, shows signatures of infall in a hot dense core (T ~ 170 K, nH2 ~ 5 × 107 cm-3), based on asymmetric spectral profiles (skewed toward the blueshifted component) in optically thick emission lines of NH3. The relatively large amount of hot molecular gas available for accretion (~20 M? within about half an arcsecond or 2500 AU), along with strong outflow and maser activity, indicates that the main accretion center in the W51e2 complex is W51e2-E rather than W51e2-W. Finally, W51e2-NW and W51e8, although less dense (nH2 ~ 2 × 107 cm-3 and ~ 3 × 106 cm-3), are also hot cores (Tgas ~ 140 and 200 K) and contain a significant amount of molecular gas (Mgas ~ 30M? and ~70 M?, respectively). We speculate that they may host high-mass YSOs either at a previous evolutionary stage or with a mass that is lower than W51e2-E and W51e2-W. Conclusions: Using high-angular resolution multilevel imaging of highly excited NH3 metastable lines, we characterized the physical and dynamical properties of four individual high-mass young stars forming in the W51 Main clump. The reduced NH3 data cubes as FITS files are only available at the CDS via anonymous ftp to http://cdsarc.u-strasbg.fr (ftp://130.79.128.5) or via http://cdsarc.u-strasbg.fr/viz-bin/qcat?J/A+A/589/A44
Context. This paper is the third in a series of NH3 multilevel imaging studies in well-known, high-mass star-forming regions. The main goal is to characterize kinematics and physical conditions of (hot and dense) circumstellar molecular gas around O-type young stars. Aims: We want to map at subarcsecond resolution highly excited inversion lines of NH3 in the high-mass star-forming region W51 Main (distance = 5.4 kpc), which is an ideal target to constrain theoretical models of high-mass star formation. Methods: Using the Karl Jansky Very Large Array (JVLA), we mapped the hot and dense molecular gas in W51 Main with ~0.2 arcsec-0.3 arcsec angular resolution in five metastable (J = K) inversion transitions of ammonia (NH3): (J,K) = (6, 6), (7, 7), (9, 9), (10, 10), and (13, 13). These lines arise from energy levels between ~400 K and ~1700 K above the ground state. We also made maps of the (free-free) continuum emission at frequencies between 25 and 36 GHz. Results: We have identified and characterized two main centers of high-mass star formation in W51 Main, which excite hot cores and host one or multiple high-mass young stellar objects (YSOs) at their centers: the W51e2 complex and the W51e8 core (~6'' southward of W51e2). The former breaks down into three further subcores: W51e2-W, which surrounds the well-known hypercompact (HC) HII region, where hot NH3 is observed in absorption, and two additional dusty cores, W51e2-E (~0.8 arcsec to the East) and W51e2-NW (~1'' to the North), where hot NH3 is observed in emission. The velocity maps toward the HC HII region show a clear velocity gradient along the east-west in all lines. The gradient may indicate rotation, although any Keplerian motion must be on smaller scales ( is observed in emission. The velocity maps toward the HC HII region show a clear velocity gradient along the east-west in all lines. The gradient may indicate rotation, although any Keplerian motion must be on smaller scales (⊙, assuming [NH3]/[H2] = 10-7) with respect to the mass of the central YSO estimated from radio luminosity (>20 M⊙), both indicate that the central YSO has already accreted most of its final mass. On the other hand, the nearby W51e2-E, while not showing evidence of rotation, shows signatures of infall in a hot dense core (T ~ 170 K, nH2 ~ 5 × 107 cm-3), based on asymmetric spectral profiles (skewed toward the blueshifted component) in optically thick emission lines of NH3. The relatively large amount of hot molecular gas available for accretion (~20 M⊙ within about half an arcsecond or 2500 AU), along with strong outflow and maser activity, indicates that the main accretion center in the W51e2 complex is W51e2-E rather than W51e2-W. Finally, W51e2-NW and W51e8, although less dense (nH2 ~ 2 × 107 cm-3 and ~ 3 × 106 cm-3), are also hot cores (Tgas ~ 140 and 200 K) and contain a significant amount of molecular gas (Mgas ~ 30M⊙ and ~70 M⊙, respectively). We speculate that they may host high-mass YSOs either at a previous evolutionary stage or with a mass that is lower than W51e2-E and W51e2-W. Conclusions: Using high-angular resolution multilevel imaging of highly excited NH3 metastable lines, we characterized the physical and dynamical properties of four individual high-mass young stars forming in the W51 Main clump. The reduced NH3 data cubes as FITS files are only available at the CDS via anonymous ftp to http://cdsarc.u-strasbg.fr (ftp://130.79.128.5) or via http://cdsarc.u-strasbg.fr/viz-bin/qcat?J/A A/589/A44
