arXiv subjects

Tuomas Kangas

Publications and source records attributed to Tuomas Kangas.

21 records · Page 2Linked to original sources

arXiv2018

SN 2017dio: a type-Ic supernova exploding in a hydrogen-rich circumstellar medium

SN 2017dio shows both spectral characteristics of a type-Ic supernova (SN) and signs of a hydrogen-rich circumstellar medium (CSM). Prominent, narrow emission lines of H and He are superposed on the continuum. Subsequent evolution revealed that the SN ejecta are interacting with the CSM. The initial SN Ic identification was confirmed by removing the CSM interaction component from the spectrum and comparing with known SNe Ic, and reversely, adding a CSM interaction component to the spectra of known SNe Ic and comparing them to SN 2017dio. Excellent agreement was obtained with both procedures, reinforcing the SN Ic classification. The light curve constrains the pre-interaction SN Ic peak absolute magnitude to be around $M_g = -17.6$ mag. No evidence of significant extinction is found, ruling out a brighter luminosity required by a SN Ia classification. These pieces of evidence support the view that SN 2017dio is a SN Ic, and therefore the first firm case of a SN Ic with signatures of hydrogen-rich CSM in the early spectrum. The CSM is unlikely to have been shaped by steady-state stellar winds. The mass loss of the progenitor star must have been intense, $\dot{M} \sim 0.02$ $(ε_{Hα}/0.01)^{-1}$ $(v_\textrm{wind}/500$ km s$^{-1}$) $(v_\textrm{shock}/10 000$ km s$^{-1})^{-3}$ $M_\odot$~yr$^{-1}$, peaking at a few decades before the SN. Such a high mass loss rate might have been experienced by the progenitor through eruptions or binary stripping.

astro-ph.SR↗

arXiv2014

High Density Circumstellar Interaction in the Luminous Type IIn SN 2010jl: The first 1100 days

HST and ground based observations of the Type IIn SN 2010jl are analyzed, including photometry, spectroscopy in the ultraviolet, optical and NIR bands, 26-1128 days after first detection. At maximum the bolometric luminosity was $\sim 3\times10^{43}$ erg/s and even at 850 days exceeds $10^{42}$ erg/s. A NIR excess, dominating after 400 days, probably originates in dust in the circumstellar medium (CSM). The total radiated energy is $> 6.5\times10^{50}$ ergs, excluding the dust component. The spectral lines can be separated into one broad component due to electron scattering, and one narrow with expansion velocity $\sim 100$ km/s from the CSM. The broad component is initially symmetric around zero velocity but becomes blueshifted after $\sim 50$ days, while remaining symmetric about a shifted centroid velocity. Dust absorption in the ejecta is unlikely to explain the line shifts, and we attribute the shift instead to acceleration by the SN radiation. From the optical lines and the X-ray and dust properties, there is strong evidence for large scale asymmetries in the CSM. The ultraviolet lines indicate CNO processing in the progenitor, while the optical shows a number of narrow coronal lines excited by the X-rays. The bolometric light curve is consistent with a radiative shock in an $r^{-2}$ CSM with a mass loss rate of $\sim 0.1$ M_sun/yr. The total mass lost is $> 3$ M_sun. These properties are consistent with the SN expanding into a CSM characteristic of an LBV progenitor with a bipolar geometry. The apparent absence of nuclear processing is attributed to a CSM still opaque to electron scattering.

astro-ph.HE↗

arXiv2013

Spatial distributions of core-collapse supernovae in infrared-bright galaxies

We have measured the correlation between the locations of core-collapse supernovae (CCSNe) and host galaxy light in the Ha line, near ultraviolet (NUV), R-band and Ks-band to constrain the progenitors of CCSNe using pixel statistics. Our sample consists of 86 CCSNe in 57 infrared (IR)-bright galaxies, of which many are starbursts and ten are luminous infrared galaxies (LIRGs). We also analyse the radial distribution of CCSNe in these galaxies, and determine power-law and exponential fits to CCSN surface density profiles. To probe differences between the SNe of these galaxies and normal spiral galaxies, our results were compared to previous studies with samples dominated by normal spiral galaxies where possible. We obtained a normalised scale length of 0.23^{+0.03}_{-0.02} R_25 for the CCSN surface density in IR-bright galaxies; less than that derived for CCSNe in a sample dominated by normal spiral galaxies (0.29 \pm 0.01). This reflects a more centrally concentrated population of massive stars in IR-bright galaxies. Furthermore, this centralisation is dominated by a central excess of type Ibc/IIb SNe. This may be due to a top-heavy initial mass function and/or an enhanced close binary fraction in regions of enhanced star formation. Type Ic SNe are most strongly correlated with Ha light and NUV-bright regions, reflecting the shortest lifetime and thus highest mass for type Ic progenitors. Previous studies with samples dominated by normal spiral galaxies have indicated a lower Ibc-Ha correlation than our results do, which may be due to the central excess of type Ibc/IIb SNe in our sample. The difference between types II and Ib is minimal, suggesting that progenitor mass is not the dominant factor in determining if a SN is of type Ib or II. Similar differences in correlation can be seen in the Ks-band, with type Ibc/IIb SNe tracing the Ks-band light better than type II in our sample.

astro-ph.HE↗