Abstract
Over the past two decades, the discovery of an extraordinarily bright class of transient, termed "superluminous" supernovae, has pushed the boundaries of standard SN models. While Type II SLSNe may represent the limit of the ordinary interacting SN population, their enormous kinetic energy scale surpasses what is typical of core-collapse SNe. Type I SLSNe have proven even more difficult to explain, possessing light curves that are too luminous to be powered by the decay of radioactive material synthesized in the explosion. The leading SLSN model proposes that a highly magnetized, rapidly spinning pulsar (a "magnetar") powers the extreme light curves of these events. There remain a number of open questions related to this model, however, the answers to which may help to establish the extent of the connection between SLSNe and other high energy transients. This dissertation will aim to address some of these open questions, investigating the limitations of the magnetar model for SLSNe, and exploring aspects of interacting SNe at multiple luminosity scales.
In Chapter 2 I discuss the compilation of a set of 238 SLSN light curves, which stands among the largest to date. I discuss the strengths and weaknesses of the leading magnetar-driven model for SLSNe, before fitting a simple parameterization of the magnetar model to our SLSN sample using MOSFiT. I discuss the derived parameter values, noting the common correlations and degeneracies that are seen, many of which can be tied to the limitations of the magnetar model parameterization. Our fitting results exhibit modest discrepancies when compared to previous works, reinforcing the importance of carefully chosen priors.
In Chapter 3 I expand upon the modeling introduced in Chapter 2, modifying the simple magnetar model within MOSFiT to treat the initial SN explosion energy as a free parameter. Although a number of statistical and systematic uncertainties make the explosion energy difficult to constrain, around 10% of the events in our sample are found to be consistent with a value surpassing the 10^51 erg limit of the standard delayed neutrino explosion mechanism. I discuss the factors that push our fits in this direction, with many high explosion energy events exhibiting unusually high ejecta velocities. Some events, however, rely more subtly on the interplay between the ejecta density distribution and the injected magnetar energy, illuminating the key factors that lead to the efficient conversion of injected energy into radiation that is needed for a superluminous light curve. I also discuss an alternative CSM interaction-driven model for SLSNe, noting how the known limitations of the most commonly used parameterization are likely to affect the modeling results, and proposing a couple of key first steps towards the future modeling of CSM-driven SLSNe.
In Chapter 4 I pivot further towards interacting SNe IIn, discussing the debate over the origin of the IR excess seen in the light curve of SN 2010jl, and considering the arguments for its origin either in emission from pre-existing dust in the CSM, or from newly formed dust in the ejecta-CSM interaction region. I use a combination of semi-analytic calculations and Monte Carlo radiative transfer simulations to model the IR light curve of SN 2010jl with a light echo produced by CSM dust. Building upon previous works, I find that an echo from an asymmetric distribution of CSM dust (most plausibly an inclined toroid) can fit the first 200 days of the IR light curve of SN 2010jl. I demonstrate, however, that the rise in relatively warm IR emission that is seen between 400 and 1000 days cannot be explained by a light echo, even for very small dust grains with a significantly asymmetric spatial distribution. This finding reinforces the need for more detailed modeling of the radiation produced by the ejecta-CSM interaction.
