Radiative-transfer models for dusty Type II supernovae

Abstract

Dust is expected to form on a year timescale in core-collapse supernova (SN) ejecta. Its existence is revealed through an infrared brightening, an optical dimming, or a blue-red emission-line profile asymmetry. To investigate how the dust location and amount impact observations, we computed ultraviolet-to-optical spectra of interacting and standard, noninteracting Type II SNe using state-of-the-art models – for simplicity we adopted 0.1 μm silicate grains. These models account for the full ejecta and treat both radioactive decay and shock power that arises from interaction of the ejecta with circumstellar material. In a Type IIn SN such as 1998S at one year, approximately 3×10‑4 M⊙ of dust within the dense shell reproduces the broad, asymmetric Hα profile. It causes an optical dimming of ∼2 mag (which obscures any emission from the inner, metal-rich ejecta) but, paradoxically, a more modest dimming of the ultraviolet, which originates from the outer parts of the dense shell. In Type II SNe with late-time interaction, such as SN 2017eaw, dust in the low-mass, fast outer ejecta dense shell tends to be optically thin, impacting little the optical spectrum for dust masses of order 10‑4 M⊙. In such SNe II with interaction, dust in the inner metal-rich ejecta has negligible effect on observed spectra in the ultraviolet and optical. In noninteracting SNe II, dust within the metal-rich ejecta preferentially quenches the [O I] λλ 6300, 6364 and [Ca II] λλ 7291, 7323 metal lines, biasing the emission in favor of the H-rich material which generates the Hα and Fe II emission below 5500 Å. Our model with 5×10‑4 M⊙ of dust below 2000 km s‑1 closely matches the optical spectrum of SN 1987A at 714 d. Modeling historical SNe requires one to treat both the ejecta material and the dust, as well as multiple power sources, although interaction power will generally dominate.