Pristine stars with masses between ∼140 and 260 M⊙ are theoretically predicted to die as pair‐instability supernovae. These very massive progenitors could come from Population III (Pop III) stars in the early universe. We model the light curves and spectra of pair‐instability supernovae over a range of masses and envelope structures. At redshifts of reionization z≥ 6, we calculate the rates and detectability of pair‐instability and core‐collapse supernovae, and show that with the James Webb Space Telescope it is possible to determine the contribution of Pop III and Pop II stars towards reionization by constraining the stellar initial mass function at that epoch using these supernovae. We also find the rates of Type Ia supernovae, and show that they are not rare during reionization, and can be used to probe the mass function at 4–8 M⊙. If the budget of ionizing photons was dominated by contributions from top‐heavy Pop III stars, we predict that the bright end of the galaxy luminosity function will be contaminated by pair‐instability supernovae.

Stars with helium cores between ∼64 and 133 M⊙ are theoretically predicted to die as pair‐instability supernovae. This requires very massive progenitors, which are theoretically prohibited for Pop II/I stars within the Galactic stellar mass limit due to mass‐loss via line‐driven winds. However, the runaway collision of stars in a dense, young star cluster could create a merged star with sufficient mass to end its life as a pair‐instability supernova, even with enhanced mass‐loss at non‐zero metallicity. We show that the predicted rate from this mechanism is consistent with the inferred volumetric rate of roughly ∼2 × 10−9 Mpc−3 yr−1 of the two observed pair‐instability supernovae, SN 2007bi and PTF 10nmn, neither of which has metal‐free host galaxies. Contrary to prior literature, only pair‐instability supernovae at low redshifts z < 2 will be observable with the Large Synoptic Survey Telescope. We estimate that the telescope will observe ∼102 such events per year that originate from the collisional runaway mergers in clusters.

For supernova powered by the conversion of kinetic energy into radiation due to the interactions of the ejecta with a dense circumstellar shell, we show that there could be X-ray analogues of optically super-luminous SNe with comparable luminosities and energetics. We consider X-ray emission from the forward shock of SNe ejecta colliding into an optically-thin CSM shell, derive simple expressions for the X-ray luminosity as a function of the circumstellar shell characteristics, and discuss the different regimes in which the shock will be radiative or adiabatic, and whether the emission will be dominated by free-free radiation or line-cooling. We find that even with normal supernova explosion energies of 10 51 erg, there exists CSM shell configurations that can liberate a large fraction of the explosion energy in X-rays, producing unabsorbed X-ray luminosities approaching 10 44 erg s −1 events lasting a few months, or even 10 45 erg s −1 flashes lasting days. Although the large column density of the circumstellar shell can absorb most of the flux from the initial shock, the most luminous events produce hard X-rays that are less susceptible to photoelectric absorption, and can counteract such losses by completely ionizing the intervening material. Regardless, once the shock traverses the entire circumstellar shell, the full luminosity could be available to observers.If va ≫ v b , and the lost kinetic energy is converted to radiation with efficiency α, then the total radiated energy will shock energy in optically thick CSM, i.e. the shock breakout, which can also produce X-rays; see Section 6. c 2013 RAS

Current surveys are underway to utilize gravitational lensing by galaxy clusters with Einstein radii > 35 ′′ in the search for the highest redshift galaxies. Associated supernova from the epoch of reionization would have their fluxes boosted above the detection threshold, extending their duration of visibility. We predict that the James Webb Space Telescope (JWST) will be able to discover lensed core-collapse supernovae at redshifts exceeding z = 7-8.

We examine the possibility that very massive stars greatly exceeding the commonly adopted stellar mass limit of 150M ⊙ may be present in young star clusters in the local universe. We identify ten candidate clusters, some of which may host stars with masses up to 600M ⊙ formed via runaway collisions. We estimate the probabilities of these very massive stars being in eclipsing binaries to be 30%. Although most of these systems cannot be resolved at present, their transits can be detected at distances of 3 Mpc even under the contamination of the background cluster light, due to the large associated luminosities ∼ 10 7 L ⊙ and mean transit depths of ∼ 10 6 L ⊙ . Discovery of very massive eclipsing binaries would flag possible progenitors of pair-instability supernovae and intermediate-mass black holes.