Prof Patrick Roche

Abstract:

Most massive stars end their lives in core-collapse supernova explosions and enrich the interstellar medium with explosively nucleosynthesized elements. Following core collapse, the explosion is subject to instabilities as the shock propagates outward through the progenitor star. Observations of the composition and structure of the innermost regions of a core-collapse supernova provide a direct probe of the instabilities and nucleosynthetic products. SN 1987A in the Large Magellanic Cloud is one of very few supernovae for which the inner ejecta can be spatially resolved but are not yet strongly affected by interaction with the surroundings. Our observations of SN 1987A with the Atacama Large Millimeter/submillimeter Array are of the highest resolution to date and reveal the detailed morphology of cold molecular gas in the innermost regions of the remnant. The 3D distributions of carbon and silicon monoxide (CO and SiO) emission differ, but both have a central deficit, or torus-like distribution, possibly a result of radioactive heating during the first weeks ("nickel heating"). The size scales of the clumpy distribution are compared quantitatively to models, demonstrating how progenitor and explosion physics can be constrained.

Abstract:

We present new calculations of the CAT3D clumpy torus models, which now include a more physical dust sublimation model as well as AGN anisotropic emission. These new models allow graphite grains to persist at temperatures higher than the silicate dust sublimation temperature. This produces stronger near-infrared emission and bluer mid-infrared (MIR) spectral slopes. We make a statistical comparison of the CAT3D model MIR predictions with a compilation of sub-arcsecond resolution ground-based MIR spectroscopy of 52 nearby Seyfert galaxies (median distance of 36 Mpc) and 10 quasars. We focus on the AGN MIR spectral index αMIR and the strength of the 9.7 μm silicate feature SSil. As with other clumpy torus models, the new CAT3D models do not reproduce the Seyfert galaxies with deep silicate absorption (SSil < −1). Excluding those, we conclude that the new CAT3D models are in better agreement with the observed αMIR and SSil of Seyfert galaxies and quasars. We find that Seyfert 2 are reproduced with models with low photon escape probabilities, while the quasars and the Seyfert 1-1.5 require generally models with higher photon escape probabilities. Quasars and Seyfert 1-1.5 tend to show steeper radial cloud distributions and fewer clouds along an equatorial line-of-sight than Seyfert 2. Introducing AGN anisotropic emission besides the more physical dust sublimation models alleviates the problem of requiring inverted radial cloud distributions (i.e., more clouds towards the outer parts of the torus) to explain the MIR spectral indices of type 2 Seyferts.

Abstract:

We report the first molecular line survey of Supernova 1987A in the millimetre wavelength range. In the Atacama Large Millimeter/submillimeter Array (ALMA) 210–300 and 340–360 GHz spectra, we detected cold (20–170 K) CO, ²⁸SiO, HCO⁺ and SO, with weaker lines of ²⁹SiO from ejecta. This is the first identification of HCO⁺ and SO in a young supernova remnant. We find a dip in the J = 6–5 and 5–4 SiO line profiles, suggesting that the ejecta morphology is likely elongated. The difference of the CO and SiO line profiles is consistent with hydrodynamic simulations, which show that Rayleigh–Taylor instabilities cause mixing of gas, with heavier elements much more disturbed, making more elongated structure. We obtained isotopologue ratios of ²⁸SiO/²⁹SiO > 13, ²⁸SiO/³⁰SiO > 14 and ¹²CO/¹³CO > 21, with the most likely limits of ²⁸SiO/²⁹SiO >128, ²⁸SiO/³⁰SiO >189. Low ²⁹Si and ³⁰Si abundances in SN 1987A are consistent with nucleosynthesis models that show inefficient formation of neutron-rich isotopes in a low-metallicity environment, such as the Large Magellanic Cloud. The deduced large mass of HCO⁺ (∼5 × 10⁻⁶ M_⊙) and small SiS mass (<6 × 10⁻⁵ M_⊙) might be explained by some mixing of elements immediately after the explosion. The mixing might have caused some hydrogen from the envelope to sink into carbon- and oxygen-rich zones after the explosion, enabling the formation of a substantial mass of HCO⁺. Oxygen atoms may have penetrated into silicon and sulphur zones, suppressing formation of SiS. Our ALMA observations open up a new window to investigate chemistry, dynamics and explosive nucleosynthesis in supernovae.

Abstract:

We investigate the evolutionary connection between local infrared (IR)-bright galaxies (logLIR ≥11.4 Lʘ) and quasars. We use high-angular resolution (∼0.3–0.4 arcsec∼few hundred parsecs) 8–13µm ground-based spectroscopy to disentangle the active galactic nuclei (AGN) id-IR properties from those of star formation. The comparison between the nuclear 11.3µm polycyclic aromatic hydrocarbon feature emission and that measured with Spitzer/Spitzer Infrared Spectrograph indicates that the star formation is extended over a few kpc in the IRbright galaxies. The AGN contribution to the total IR luminosity of IR-bright galaxies is lower than in quasars. Although the dust distribution is predicted to change as IR-bright galaxies evolve to IR-bright quasars and then to optical quasars, we show that the AGN mid-IR emission of all the quasars in our sample is not significantly different. In contrast, the nuclear emission of IR-bright galaxies with low AGN contributions appears more heavily embedded in dust although there is no clear trend with the interaction stage or projected nuclear separation. This suggests that the changes in the distribution of the nuclear obscuring material may be taking place rapidly and at different interaction stages washing out the evidence of an evolutionary path. When compared to normal AGN, the nuclear star formation activity of quasars appears to be dimming, whereas it is enhanced in some IR-bright nuclei, suggesting that the latter are in an earlier star formation-dominated phase.