Context. The progenitors of many Type II supernovae have been observationally identified but the search for Type Ibc supernova (SN Ibc) progenitors has thus far been unsuccessful, despite the expectation that they are luminous Wolf-Rayet (WR) stars. Aims. We investigate how the evolution of massive helium stars affects their visual appearances, and discuss the implications for the detectability of SN Ibc progenitors.Methods. Evolutionary models of massive helium stars are analysed and their properties compared to Galactic WR stars. Results. Massive WR stars that rapidly lose their helium envelopes through stellar-wind mass-loss end their lives when their effective temperatures -related to their hydrostatic surfaces -exceed about 150 kK. At their pre-supernova stage, their surface properties resemble those of hot Galactic WR stars of WO sub-type. These are visually faint with narrow-band visual magnitudes M v = −1.5 · · · −2.5, despite their high bolometric luminosities (log L/L = 5.6 · · · 5.7), compared to the bulk of Galactic WR stars (M v < −4). In contrast, relatively low-mass helium stars that retain a thick helium envelope appear fairly bright in optical bands, depending on the final masses and the history of the envelope expansion during the late evolutionary stages. Conclusions. We conclude that SNe Ibc observations have so far not provided strong constraints on progenitor bolometric luminosities and masses, even with the deepest searches. We also argue that Ic progenitors are more challenging to identify than Ib progenitors in any optical images.

Context. So-called superluminous supernovae have been recently discovered in the local Universe. It appears possible that some of them originate in stellar explosions induced by the pair instability mechanism. Recent stellar evolution models also predict pair instability supernovae from very massive stars at fairly high metallicities (i.e., Z ∼ 0.004). Aims. We provide supernova models and synthetic light curves for two progenitor models, a 150 M red supergiant and a 250 M yellow supergiant at a metallicity of Z = 0.001, for which the evolution from the main sequence to collapse and the initiation of the pair instability supernova itself has been previously computed in a realistic and self-consistent way. Methods. We use the radiation hydrodynamics code STELLA to describe the supernova evolution of both models in a time frame of about 500 days. Results. We describe the shock-breakout phases of both supernovae, which are characterized by higher luminosity, longer duration, and a lower effective temperature than those of ordinary Type IIP supernovae. We derive the bolometric, as well as the U, B, V, R, and I, light curves of our pair instability supernova models, which show a long-lasting plateau phase with maxima at M bol −19.3 mag and −21.3 mag for our lower and higher mass models, respectively. While we do not produce synthetic spectra, we also describe the photospheric composition and velocity as a function of time. Conclusions. We conclude that the light curve of the explosion of our initially 150 M star resembles those of relatively bright type IIP supernovae, whereas its photospheric velocity at early times is somewhat lower. Its 56 Ni mass of 0.04 M also falls well into the range found in ordinary core collapse supernovae. The light curve and photospheric velocity of our 250 M models has a striking resemblance to that of the superluminous SN 2007bi, strengthening its interpretation as pair instability supernova. We conclude that pair instability supernovae may occur more frequently in the local universe than previously assumed.

Context. Both recent observations and stellar evolution models suggest that pair-instability supernovae (PISNe) could occur in the local Universe, at metallicities below < ∼ Z /3. Previous PISN models were mostly produced at very low metallicities in the context of the early Universe. Aims. We present new PISNe models at a metallicity of Z = 0.001, which are relevant for the local Universe. Methods. We took previously published self-consistent stellar evolutionary models of pair-instability progenitors with initial masses of 150 M and 250 M at metallicity of Z = 0.001 and followed the evolution of these models through the supernova explosions, using a hydrodynamics stellar evolution code with an extensive nuclear network including 200 isotopes. Results. In both models the stars explode as PISNe without leaving a compact stellar remnant. Our models produce a nucleosynthetic pattern that is generally similar to that of Population III PISN models, which is mainly characterized by the production of large amounts of α-elements and a strong deficiency of the odd-charged elements. However, the odd-even effect in our models is significantly weaker than that found in Population III models. The comparison with the nucleosynthetic yields from core-collapse supernovae at a similar metallicity (Z = 0.002) indicates that PISNe could have strongly influenced the chemical evolution below Z ≈ 0.002, assuming a standard initial mass function. The odd-even effect is predicted to be most prominent for the intermediatemass elements between silicon and calcium. Conclusions. With future observations of chemical abundances in Population II stars, our result can be used to constrain the number of PISNe that occurred during the past evolution of our Galaxy.

With an increasing number of superluminous supernovae (SLSNe) discovered the question of their origin remains open and causes heated debates in the supernova community. Currently, there are three proposed mechanisms for SLSNe: (1) pair-instability supernovae (PISN), (2) magnetar-driven supernovae, and (3) models in which the supernova ejecta interacts with a circumstellar material ejected before the explosion. Based on current observations of SLSNe, the PISN origin has been disfavoured for a number of reasons. Many PISN models provide overly broad light curves and too reddened spectra, because of massive ejecta and a high amount of nickel. In the current study we re-examine PISN properties using progenitor models computed with the GENEC code. We calculate supernova explosions with FLASH and light curve evolution with the radiation hydrodynamics code STELLA. We find that high-mass models (200 M⊙ and 250 M⊙) at relatively high metallicity (Z = 0.001) do not retain hydrogen in the outer layers and produce relatively fast evolving PISNe Type I and might be suitable to explain some SLSNe. We also investigate uncertainties in light curve modelling due to codes, opacities, the nickel-bubble effect and progenitor structure and composition.