Summary: | This thesis studies type Ia supernovae (SNe Ia) discovered by the Palomar Transient Factory (PTF). We use this large, high-quality imaging and spectroscopic dataset to analyse the very earliest part of the SN Ia light curve, probing the ejecta structure, the postmaximum light curve, the 56Ni distribution, and the utility of SNe Ia as standard candles at different wavelengths. The high cadence of PTF makes it capable of studying SNe Ia at very early times, as they rise just after explosion. In this thesis we use 18 SNe Ia from PTF and the La Silla-QUEST variability survey (LSQ) to measure the time between first light, t0, and maximum light (t_rise), and rise index (n), where f = (t - t0)n. We find that n shows significant variation (1:48 ≤ n ≤ 3:70), and has a mean value n̄ = 2.44 ± 0.13, inconsistent with a simple fireball model (n = 2) at 3σ. We calculate an average t_rise of our sample t_rise = 18:98 ± 0:54 days, and find that the variation seen in t_rise and n is principally driven at the very earliest epochs. We use a further sample of 31 PTF SNe Ia to explore the diversity of the post-maximum light curve in R-band, as the emission at these phases is linked to the same driver of variation - the distribution of 56Ni through the ejecta. By modelling the shoulder on the light curve we quantify the strength and timing of this feature. We find that the timing of the additional emission is correlated with light curve width, but that the duration is not. We also find that the prominence of the gaussian, relative to its peak, is inversely correlated with the light-curve width. We interpret this as the impact of intermediate mass element spectral features dominating over those of iron peak elements. Having explored the physics that governs these objects, we turn to the primary motivation for studying SNe Ia - their use as standardisable candles. We consider 60 SNe Ia in three filters, B, R and i, constructing hubble diagrams for each. We find that the magnitude of the corrections for light curve width and SN colour decreases as a function of wavelength, and that SNe Ia in i-band do not need correction for light curve width. We also find that the colour law observed for our sample is not consistent with the standard Milky-Way extinction law, with RV = 3.1. Finally, we find that SNe Ia in i are superior standardisable candles to B-band, and that they are overall the best in R. This sample is well placed to anchor future high-redshift rest-frame i-band cosmological studies.
|