Understanding the diversity of Type Ia supernovae from early to late times

Abstract

Type Ia Supernovae (SNe Ia) have been pivotal in the formation of the 21st Century view of the universe due to their involvement in the discovery of the accelerated expansion of the universe. However, an all-encompassing explanation of their progenitor systems and explosion mechanisms is lacking, bringing into question their reliability as cosmic distance indicators. In order to address the long-standing questions in SN Ia physics, I begin by studying the early light curves of SNe Ia (0-20 d post-explosion). These early epochs can inform us about outer regions of the ejecta, which can shed light on the explosion mechanism and progenitor system. Using a homogeneous sample of 115 SNe Ia provided by the Zwicky Transient Facility (ZTF) I fit SN Ia light curves with Chandrasekhar-mass explosion models. I show that ~67 per cent of the sample can be reproduced by models with extended 56Ni distributions, implying at least some level of mixing in the SN ejecta. Flux excesses in the first few days after explosion could be key to unravelling the progenitor scenarios of SNe Ia. I find six SNe Ia showing a flux excess in the sample of 115 SNe Ia. I performed an efficiency analysis, and in the lowest redshift bin (z < 0.07) I find a detection efficiency of 57 per cent, and an intrinsic rate of bumps of 18 +/- 11 per cent. The duration and colour of the bumps in the sample are consistent with interactions with a non-degenerate companion or dense CSM, and nickel-clump models, and I note that spectroscopic information would be required in future studies to distinguish between these scenarios. Next, I explore the secondary maximum, which occurs 15-35 d after maximum light and contains crucial information about the conditions of the ejecta. I study a sample of 893 ZTF SNe Ia with well-sampled r and i-band light curves and fit them using Gaussian Processes. From these fits I calculate the time of the onset of the secondary maximum as well as the strength of the secondary maximum. I find that brighter, slower evolving SNe Ia have a later and stronger secondary maximum. I present the distribution of transparency timescales for the sample and compare it to predictions from various explosion models, and find that 97 per cent of the sample is consistent with the predictions from double detonation explosion models. I find the transparency timescale to be best fit by a bimodal Gaussian, with the lower transparency timescale component originating predominantly from SNe Ia in red local environments ((g - z)local > 1.0). Finally, I present an investigation of a sample of 24 nearby (z < 0.025) SNe Ia, with NIR data during the near-infrared (NIR) plateau phase (between 70-500 d), and confirm that the plateau only occurs in the J and H bands, whilst the K band continues to decline. It is shown that SNe Ia with broader optical light curves at peak tend to have a higher average brightness on the plateau in J and H, and SNe Ia that are more luminous at peak also show a steeper decline during the plateau phase in H. I compare the data to state-of-the-art radiative transfer models of nebular SNe Ia in the near-infrared, and find good agreement with a sub-Mch model. An analysis of the spectral evolution during the plateau demonstrates that the ratio of [Fe ii] to [Fe iii] determines the slope of the plateau.