Superluminous supernovae: from individual extremes to population trends

Abstract

Superluminous supernovae (SLSNe) are among the most luminous stellar explosions known, offering a unique insight into the lives and deaths of the most massive stars. These explosions require additional energy sources beyond the decay of 56Ni, with central engines such as rapidly rotating magnetars, or strong circumstellar medium (CSM) interaction, among the leading candidates. Despite significant progress over the past decade, their progenitor systems, explosion mechanisms, and powering sources remain subjects of debate. The motivation of this thesis is to investigate the diverse nature of SLSNe through detailed studies of unusual individual events and analyses of large populations. The aim is to constrain their progenitor properties, explosion mechanisms, and observable diversity.

The hydrogen-poor SLSN 2019szu provides a striking case study. The detection of pre-maximum forbidden [O II] emission lines (among the earliest such features ever observed) revealed ∼0.25 M⊙ of O-rich, low-density material ejected less than 120 days before explosion. This mass loss can naturally be explained by pulsational pair-instability (PPI) ejections. The PPI scenario also accounts for the observed spectral signatures for interaction and the pre-explosion light curve plateau through
shell–shell collisions. This work suggests that early-time forbidden lines may serve as a diagnostic for identifying future PPI candidates.

This thesis also presents the largest uniformly processed sample of SLSN spectra to date, nearly 1000 spectra for 234 events. Principal component analysis finds no statistical evidence for distinct subclasses within the population, but is able to identify outlier events with evidence for energy input from interaction. Measurements of line velocities and velocity gradients indicate line-forming regions close to the photosphere and ejecta in homologous expansion. These findings are consistent with central engine models. This homogeneous dataset will facilitate future machine learning applications to SLSNe.

Finally, I explore the ultraviolet (UV) properties of SLSNe through the unusual case of SN 2023taz, whose red UV colours and deep Mg II absorption point to greater intrinsic diversity in the UV than previously understood. As upcoming facilities push SLSN discoveries to higher redshift where rest-frame UV may be the only accessible wavelength range, understanding this diversity at low redshift will be critical. Coordinated UV–optical campaigns will be essential for developing new classification schemes that can be applied to the next decade of transient surveys.

Date of Award	Jul 2026
Original language	English
Awarding Institution	Queen's University Belfast
Supervisor	Matt Nicholl (Supervisor) & Stuart Sim (Supervisor)