Blue Supergiant Progenitor Research Articles

The Influence of Stellar Rotation in Binary Systems on Core-collapse Supernova Progenitors and Multimessenger Signals

The detailed structure of core-collapse supernova progenitors is crucial for studying supernova explosion engines and the corresponding multimessenger signals. In this paper, we investigate the influence of stellar rotation on binary systems consisting of a 30M ⊙ donor star and a 20M ⊙ accretor using the MESA stellar evolution code. We find that through mass transfer in binary systems, fast-rotating red- and blue-supergiant progenitors can be formed within a certain range of the initial orbital periods, although the correlation is not linear. We also find that even with the same initial mass ratio of the binary system, the resulting final masses of the collapsars, the iron core masses, the compactness parameters, and the final rotational rates can vary widely and are sensitive to the initial orbital periods. For instance, the progenitors with strong convection form a thinner Si shell and a wider O shell compared to those in single-star systems. In addition, we conduct 2D self-consistent core-collapse supernova simulations with neutrino transport for these rotating progenitors derived from binary stellar evolution. We find that the neutrino and gravitational-wave signatures of these binary progenitors could exhibit significant variations. Progenitors with larger compactness parameters produce more massive proto-neutron stars, have higher mass accretion rates, and emit brighter neutrino luminosity and louder gravitational emissions. Finally, we observe stellar-mass black hole formation in some of our failed exploding models.

Predicting the heaviest black holes below the pair instability gap

ABSTRACT Traditionally, the pair instability (PI) mass gap is located between 50 and 130 M⊙, with stellar mass black holes (BHs) expected to ‘pile up’ towards the lower PI edge. However, this lower PI boundary is based on the assumption that the star has already lost its hydrogen (H) envelope. With the announcement of an ‘impossibly’ heavy BH of 85 M⊙ as part of GW 190521 located inside the traditional PI gap, we realized that blue supergiant (BSG) progenitors with small cores but large hydrogen envelopes at low metallicity (Z) could directly collapse to heavier BHs than had hitherto been assumed. The question of whether a single star can produce such a heavy BH is important, independent of gravitational wave events. Here, we systematically investigate the masses of stars inside the traditional PI gap by way of a grid of 336 detailed mesa stellar evolution models calculated across a wide parameter space, varying stellar mass, overshooting, rotation, semiconvection, and Z. We evolve low Z stars in the range 10−3 < Z/Z⊙ < ZSMC, making no prior assumption regarding the mass of an envelope, but instead employing a wind mass-loss recipe to calculate it. We compute critical carbon–oxygen and helium core masses to determine our lower limit to PI physics, and we provide two equations for Mcore and Mfinal that can also be of use for binary population synthesis. Assuming the H envelope falls into the BH, we confirm the maximum BH mass below PI is MBH ≃ 93.3 M⊙. Our grid allows us to populate the traditional PI gap, and we conclude that the distribution of BHs above the traditional boundary is not solely due to the shape of the initial mass function, but also to the same stellar interior physics (i.e. mixing) that which sets the BH maximum.

The Formation of the Stripped-envelope Type IIb Supernova Progenitors: Rotation, Metallicity, and Overshooting

Type IIb supernovae (SNe) are believed to originate from core-collapse progenitors having kept only a very thin hydrogen envelope. We aim to explore how some physical factors, such as rotation, metallicity, overshooting, and the initial orbital period in binaries significantly affect the Roche lobe overflow and the formation of Type IIb SNe (also denoted SNe IIb). It is found that binaries are the main channel that are capable of producing Type IIb SNe progenitors in the mass range for initial masses below 20 M ⊙. The formation of Type IIb SNe progenitors is extremely sensitive to the initial orbital period. A less massive hydrogen envelope mass indicates a smaller radius and higher effective temperatures, and vice versa. Binary systems with initial periods between 300 and 720 days produce Type IIb progenitors that are a red supergiant. Those with an initial period between 50 and 300 days produce yellow supergiant progenitors, and those with initial periods shorter than 50 days blue supergiant progenitors. Both rapid rotation and larger overshooting can enlarge the carbon–oxygen core mass and lead to higher core temperature and lower central density at the precollapse phase. They are also beneficial to surface nitrogen enrichment but restrict the efficiency of the first dredge-up. SN IIb progenitors with low metallicity have smaller hydrogen envelope masses and radii than their high-metallicity counterparts. Ultrastripped binary models have systematically higher core mass fraction 12C left, which has an important influence on the compactness of Type IIb progenitors.

SN 1987A: Tracing the flux decline and spectral evolution through a comparison of SRG/eROSITA and XMM-Newton observations

Context. SN 1987A is the supernova closest to us observed in the past four centuries. It provides the unique opportunity of witnessing the birth and evolution of a supernova remnant. Monitoring the source in X-rays provides insights into the physics of the shock, the X-ray emitting plasma, and the interaction of the shock with the structures in the circumstellar medium. The source has been monitored by XMM-Newton EPIC-pn from 2007 to 2020. SRG/eROSITA also observed the source during its commissioning phase and first light in September and October 2019. Aims. We investigated the spectral and flux evolution of SN 1987A in X-rays in the last 14 yr up to November 2020 using XMM-Newton and eROSITA observations. Methods. We performed a detailed spectral analysis using a three-component plane-parallel shock model and analysed and modelled the EPIC-pn monitoring and eROSITA observations in a consistent manner. Results. This paper reports a complete and the most recent flux evolution of SN 1987A in the soft (0.5−2 keV) and hard (3−10 keV) X-ray band. The flux in the soft band flattened around 9424 days and then displayed a turnover between 10 142 and 10 493 days, after which it showed a continued decline. At about the same time, a break in the hard-band flux time evolution slope was detected. This implies that the blast wave has now passed beyond the dense structures of the equatorial ring and is expanding farther into more tenuous circumstellar medium. The temporal evolution of the normalisations of the three shock components match the results of hydrodynamical simulations well, which predict a blue supergiant progenitor scenario. The trend in recent epochs indicates that the emission caused by the forward shock after leaving the equatorial ring and by the reverse shock in the ejecta is now becoming more dominant. The elemental abundances in the hot plasma component are significantly higher than those in the cooler component, indicating an origin in the reverse shock that propagates into the ejecta.

Maximum black hole mass across cosmic time

ABSTRACT At the end of its life, a very massive star is expected to collapse into a black hole (BH). The recent detection of an 85 M⊙ BH from the gravitational wave event GW 190521 appears to present a fundamental problem as to how such heavy BHs exist above the approximately 50 M⊙ pair-instability (PI) limit where stars are expected to be blown to pieces with no remnant left. Using mesa, we show that for stellar models with non-extreme assumptions, 90–100 M⊙ stars at reduced metallicity ($Z/\mbox{ $\mathrm{Z}_{\odot }$}\le 0.1$) can produce blue supergiant progenitors with core masses sufficiently small to remain below the fundamental PI limit, yet at the same time lose an amount of mass via stellar winds that is small enough to end up in the range of an ‘impossible’ 85 M⊙ BH. The two key points are the proper consideration of core overshooting and stellar wind physics with an improved scaling of mass-loss with iron (Fe) contents characteristic for the host galaxy metallicity. Our modelling provides a robust scenario that not only doubles the maximum BH mass set by PI, but also allows us to probe the maximum stellar BH mass as a function of metallicity and cosmic time in a physically sound framework.

The infancy of core-collapse supernova remnants

ABSTRACT We present 3D hydrodynamic simulations of neutrino-driven supernovae (SNe) with the Prometheus-HotB code, evolving the asymmetrically expanding ejecta from shock breakout until they reach the homologous expansion phase after roughly one year. Our calculations continue the simulations for two red supergiant and two blue supergiant progenitors by Wongwathanarat et al., who investigated the growth of explosion asymmetries produced by hydrodynamic instabilities during the first second of the explosion and their later fragmentation by Rayleigh–Taylor instabilities. We focus on the late time acceleration and inflation of the ejecta caused by the heating due to the radioactive decay of 56Ni to 56Fe and by a new outward-moving shock, which forms when the reverse shock from the He/H-shell interface compresses the central part of the ejecta. The mean velocities of the iron-rich ejecta increase between 100 and 350 km s−1 (∼8–30 per cent), and the fastest 1 per cent of the iron accelerates by up to ∼1000 km s−1 (∼20–25 per cent). This ‘Ni-bubble effect’, known from 1D models, accelerates the bulk of the nickel in our 3D models and causes an inflation of the initially overdense Ni-rich clumps, which leads to underdense, extended fingers, enveloped by overdense skins of compressed surrounding matter. We also provide volume and surface filling factors as well as a spherical harmonics analysis to characterize the spectrum of Ni-clump sizes quantitatively. Three of our four models give volume filling factors larger than 0.3, consistent with what is suggested for SN 1987A by observations.

The Link between Hot and Cool Outflows

AbstractThe link between hot and cool stellar outflows is shown to be critical for correctly predicting the masses of the most massive black holes (BHs) below the so-called pair-instability supernova (PISN) mass gap. Gravitational Wave (GW) event 190521 allegedly hosted an “impossibly” heavy BH of 85 M⊙. Here we show how our increased knowledge of both metallicity Z and temperature dependent mass loss is critical for our evolutionary scenario of a low-Z blue supergiant (BSG) progenitor of an initially approx 100 M⊙ star to work. We show using MESA stellar evolution modelling experiments that as long as we can keep such stars above 8000 K such low-Z BSGs can avoid strong winds, and keep a very large envelope mass intact before core collapse. This naturally leads to the Cosmic Time dependent maximum BH function below the PISN gap.

Oscillation of high-energy neutrinos from choked jets in stellar and merger ejecta

We present a comprehensive study on oscillation of high-energy neutrinos from two different environments: blue supergiant progenitors that may harbor low-power gamma-ray burst (GRB) jets and neutron star merger ejecta that would be associated with short gamma-ray bursts. We incorporate the radiation constraint that gives a necessary condition for nonthermal neutrino production, and account for the time evolution of the jet, which allows us to treat neutrino oscillation in matter more accurately. For massive star progenitors, neutrino injection inside the star can lead to nonadiabatic oscillation patterns in the 1 TeV - 10 TeV and is also visible in the flavor ratio. For neutron star merger ejecta, we find a similar behavior in the 100 GeV - 10 TeV region and the oscillation may result in a $\nu_e$ excess around 1 TeV. These features, which enable us to probe the progenitors of long and short GRBs, could be seen by future neutrino detectors with precise flavor ratio measurements. We also discuss potential contributions to the diffuse neutrino flux measured by IceCube, and find parameter sets allowing choked low-power GRB jets to account for the neutrino flux in the 10 TeV - 100 TeV range without violating the existing constraints.

SN 2018hna: 1987A-like Supernova with a Signature of Shock Breakout

Abstract High-cadence ultraviolet, optical, and near-infrared photometric and low-resolution spectroscopic observations of the peculiar Type II supernova (SN) 2018hna are presented. The early-phase multiband light curves (LCs) exhibit the adiabatic cooling envelope emission following the shock breakout up to ∼ 14 days from the explosion. SN 2018hna has a rise time of ∼ 88 days in the V band, similar to SN 1987A. A 56Ni mass of ∼0.087 ± 0.004 M ⊙ is inferred for SN 2018hna from its bolometric LC. Hydrodynamical modeling of the cooling phase suggests a progenitor with a radius ∼50 R ⊙, a mass of ∼14–20 M ⊙, and an explosion energy of ∼1.7–2.9 × 1051 erg. The smaller inferred radius of the progenitor than a standard red supergiant is indicative of a blue supergiant progenitor of SN 2018hna. A subsolar metallicity (∼0.3 Z ⊙) is inferred for the host galaxy UGC 07534, concurrent with the low-metallicity environments of 1987A-like events.

Supernovae from blue supergiant progenitors: What a mess!

Supernova (SN) 1987A was classified as a peculiar Type II SN because of its long rising light curve and the persistent presence of H I lines in optical spectra. It was subsequently realized that its progenitor was a blue supergiant (BSG), rather than a red supergiant (RSG) as for normal, Type II-P, SNe. Since then, the number of Type II-pec SNe has grown, revealing a rich diversity in photometric and spectroscopic properties. In this study, using a single 15 M⊙ low-metallicity progenitor that dies as a BSG, we have generated explosions with a range of energies and 56Ni masses. We then performed the radiative transfer modeling with CMFGEN, from 1 d until 300 d after explosion for all ejecta. Our models yield light curves that rise to optical maximum in about 100 d, with a similar brightening rate, and with a peak absolute V-band magnitude spanning −14 to −16.5 mag. All models follow a similar color evolution, entering the recombination phase within a few days of explosion, and reddening further until the nebular phase. Their spectral evolution is analogous, mostly differing in line width. With this model set, we study the Type II-pec SNe 1987A, 2000cb, 2006V, 2006au, 2009E, and 2009mw. The photometric and spectroscopic diversity of observed SNe II-pec suggests that there is no prototype for this class. All these SNe brighten to maximum faster than our limited set of models, except perhaps SN 2009mw. The spectral evolution of SN 1987A conflicts with other observations in this set and conflicts with model predictions from 20 d until maximum: Hα narrows and weakens while Ba II lines strengthen faster than expected, which we interpret as signatures of clumping. SN 2000cb rises to maximum in only 20 d and shows weak Ba II lines. Its spectral evolution (color, line width and strength) is well matched by an energetic ejecta but the light curve may require strong asymmetry. The persistent blue color, narrow lines, and weak Hα absorption, seen in SN 2006V conflicts with expectations for a BSG explosion powered by 56Ni and may require an alternative power source. In contrast with theoretical expectations, observed spectra reveal a diverse behavior for lines like Ba II 6142 Å, Na I D, and Hα. In addition to diversity arising from different BSG progenitors, we surmise that their ejecta are asymmetric, clumped, and, in some cases, not solely powered by 56Ni decay.

Explosions of blue supergiants from binary mergers for SN 1987A

Based on the work of Menon & Heger (2017), we present the bolometric light curvesand spectra of the explosions of blue supergiant progenitors from binary mergers. We study SN 1987A and two other peculiar Type IIP supernovae: SN 1998A and SN 2006V. The progenitor models were produced using the stellar evolution code Keplerand then exploded using the 1D radiation hydrodynamic code Crab. The explosions of binary merger models exhibit an overall better fit to the light curve of SN 1987A than previous single star models, because of their lower helium-core masses, larger envelope masses, and smaller radii. The merger model that best matches the observational constraints of the progenitor of SN 1987A and the light curve is a model with a radius of 37 solar radii, an ejecta mass of 20.6 solar masses, an explosion energy of 1.7 Bethe, a nickel mass of 0.073 solar masses, and a nickel mixing velocity of 3,000 km/s. This model also works for SN 1998A and is comparable with earlier estimates from semi-analytic models. In the case of SN 2006V, however, a model with a radius of 150 solar radii and ejecta mass of 19.1 solar masses matches the light curve. These parameters are significantly higher than predictions from semi-analytic models for the progenitor of this supernova.

Modeling of hydrodynamic processes within high-mass X-ray binaries

AbstractHigh-mass X-ray binaries belong to the brightest objects in the X-ray sky. They usually consist of a massive O or B star or a blue supergiant while the compact X-ray emitting component is a neutron star (NS) or a black hole. Intensive matter accretion onto the compact object can take place through different mechanisms: wind accretion, Roche-lobe overflow, or circumstellar disk. In our multi-dimensional models we perform numerical simulations of the accretion of matter onto a compact companion in case of Be/X-ray binaries. Using Bondi-Hoyle-Littleton approximation, we estimate the NS accretion rate. We determine the Be/X-ray binary disk hydrodynamic structure and compare its deviation from isolated Be stars’ disk. From the rate and morphology of the accretion flow and the X-ray luminosity we improve the estimate of the disk mass-loss rate. We also study the behavior of a binary system undergoing a supernova explosion, assuming a blue supergiant progenitor with an aspherical circumstellar environment.

Mass ejection in failed supernovae: variation with stellar progenitor

We study the ejection of mass during stellar core-collapse when the stalled shock does not revive and a black hole forms. Neutrino emission during the protoneutron star phase causes a decrease in the gravitational mass of the core, resulting in an outward going sound pulse that steepens into a shock as it travels out through the star. We explore the properties of this mass ejection mechanism over a range of stellar progenitors using spherically-symmetric, time-dependent hydrodynamic simulations that treat neutrino mass loss parametrically and follow the shock propagation over the entire star. We find that all types of stellar progenitor can eject mass through this mechanism. The ejected mass is a decreasing function of the surface gravity of the star, ranging from several $M_\odot$ for red supergiants to $\sim 0.1M_\odot$ for blue supergiants and $\sim 10^{-3} M_\odot $ for Wolf-Rayet stars. We find that the final shock energy at the surface is a decreasing function of the core-compactness, and is $\lesssim 10^{47}-10^{48}$ erg in all cases. In progenitors with a sufficiently large envelope, high core-compactness, or a combination of both, the sound pulse fails to unbind mass. Successful mass ejection is accompanied by significant fallback accretion that can last from hours to years. We predict the properties of shock breakout and thermal plateau emission produced by the ejection of the outer envelope of blue supergiant and Wolf-Rayet progenitors in otherwise failed supernovae.

Type Ib and IIb Supernova Progenitors in Interacting Binary Systems

Abstract We explore properties of Type Ib and IIb SN progenitors that are produced by stable mass transfer in binary systems using a new grid of stellar evolution models from an initial primary mass in the range of at solar and Large Magellanic Cloud metallicities. We find that blue and yellow supergiant SN IIb progenitors (e.g., of SN 2008ax, SN 2011dh, and SN 2016gkg) have a hydrogen envelope mass less than about , mostly resulting from early Case B mass transfer with relatively low initial masses and/or low metallicity. Red supergiant (RSG) SN IIb progenitors (e.g., of SN 1993J, SN 2013df) are produced via late Case B mass transfers and have a more massive hydrogen envelope ( ). SN Ib progenitors are predominantly produced by early Case B mass transfers. Our models predict that SN IIb progenitors are systematically more luminous in the optical ( ) than the majority of SN Ib progenitors ( ) for our considered initial mass range. However, the optically bright progenitor of SN Ib iPTF13bvn (i.e., ) can be well explained by a relatively low-mass progenitor with a final mass of . The event rate of blue and yellow SN IIb progenitors would increase as metallicity decreases, while the event rate of SN Ib progenitors would decrease instead. By contrast, the population of RSG SN IIb progenitors would not be significantly affected by metallicity.

The quest for blue supergiants: binary merger models for the evolution of the progenitor of SN 1987A

We present the results of a detailed, systematic stellar evolution study of binary mergers for blue supergiant (BSG) progenitors of Type II supernovae. In particular, these are the first evolutionary models that can simultaneously reproduce nearly all observational aspects of the progenitor of SN 1987A, $\text{Sk}-69\,^{\circ}202$, such as its position in the HR diagram, the enrichment of helium and nitrogen in the triple-ring nebula, and its lifetime before its explosion. The merger model, based on the one proposed by Podsiadlowski 1992 et al. and Podsiadlowski 2007 et al., consists of a main sequence secondary star that dissolves completely in the common envelope of the primary red supergiant at the end of their merger. We empirically explore a large initial parameter space, such as primary masses ($15\,\text{M}_{\odot}$, $16\,\text{M}_{\odot}$, and $17\,\text{M}_{\odot}$), secondary masses ($2\,\text{M}_{\odot}$, $3\,\text{M}_{\odot}$, ..., $8\,\text{M}_{\odot}$) and different depths up to which the secondary penetrates the He core of the primary during the merger. The evolution of the merged star is continued until just before iron-core collapse and the surface properties of the 84 pre-supernova models ($16\,\text{M}_{\odot}-23\,\mathrm{M}_{\odot}$) computed have been made available in this work. Within the parameter space studied, the majority of the pre-supernova models are compact, hot BSGs with effective temperature $>12\,\text{kK}$ and radii of $30\,\text{R}_{\odot}-70\,\mathrm{R}_{\odot}$ of which six match nearly all the observational properties of $\text{Sk}-69\,^{\circ}202$.

Bolometric Light Curves of Peculiar Type II-P Supernovae

We examine the bolometric light curves of five Type II-P supernovae (SNe 1998A, 2000cb, 2006V, 2006au, and 2009E), which are thought to originate from blue supergiant progenitors like that of SN 1987A, using a new python package named SuperBoL. With this code, we calculate SNe light curves using three different common techniques common from the literature: the quasi-bolometric method, which integrates the observed photometry, the direct integration method, which additionally corrects for unobserved flux in the UV and IR, and the bolometric correction method, which uses correlations between observed colors and V-band bolometric corrections. We present here the light curves calculated by SuperBoL, along with previously published light curves, as well as peak luminosities and 56Ni yields. We find that the direct integration and bolometric correction light curves largely agree with previously published light curves, but with what we believe to be more robust error calculations, with . Peak luminosities and 56Ni masses are similarly comparable to previous work. SN 2000cb remains an unusual member of this sub-group, owing to the faster rise and flatter plateau than the other supernovae in the sample. Initial comparisons with the NLTE atmosphere code PHOENIX show that the direct integration technique reproduces the luminosity of a model supernova spectrum to ∼5% when given synthetic photometry of the spectrum as input. Our code is publicly available. The ability to produce bolometric light curves from observed sets of broadband light curves should be helpful in the interpretation of other types of supernovae, particularly those that are not well characterized, such as extremely luminous supernovae and faint fast objects.

Blue supergiant progenitors from binary mergers for SN 1987A and other Type II-peculiar supernovae

AbstractWe present results of a systematic and detailed stellar evolution study of binary mergers for blue supergiant (BSG) progenitors of Type II supernovae, particularly for SN 1987A. We are able to reproduce nearly all observational aspects of the progenitor of SN 1987A, Sk –69 °202, such as its position in the HR diagram, the enrichment of helium and nitrogen in the triple-ring nebula and its lifetime before its explosion. We build our evolutionary model based on the merger model of Podsiadlowski et al. (1992), Podsiadlowski et al. (2007) and empirically explore an initial parameter consisting of primary masses, secondary masses and different depths up to which the secondary penetrates the He core during the merger. The evolution of the post-merger star is continued until just before iron-core collapse. Of the 84 pre-supernova models (16 M⊙ − 23 M⊙) computed, the majority of the pre-supernova models are compact, hot BSGs with effective temperature >12 kK and 30 R⊙ − 70 R⊙ of which six match nearly all the observational properties of Sk –69 °202.

The quest for blue supergiants: Evolution of binary merger progenitors of Type-II peculiar supernovae and SN 1987A

AbstractWe construct stellar evolution models until core collapse using KEPLER (Woosley & Heger (2007)) to reproduce the observed signatures of the blue supergiant (BSG) progenitor of SN 1987A. This is based on the binary merger scenario proposed by Podsiadlowski (1992) and Ivanova et al. (2002). Various combinations of initial parameters for the binary components (M1=16–18 M⊙ and M2=5–10 M⊙) and their merging, successfully match the He, N/C and N/O ratios, along with the luminosity and effective temperature of the progenitor. Most of our models end their lives as BSGs. Thus we may be able to explain the origin of all Type IIP SNe that resemble SN 1987A through such mergers. We are currently working on the light curves and nuclear yields from the explosion of these models to compare them SN 1987A.

THE ROLE OF THE MAGNETOROTATIONAL INSTABILITY IN MASSIVE STARS

The magnetorotational instability (MRI) is key physics in accretion disks and is widely considered to play some role in massive-star core collapse. Models of rotating massive stars naturally develop very strong shear at composition boundaries, a necessary condition for MRI instability, and the MRI is subject to triply-diffusive destabilizing effects in radiative regions. We have used the MESA stellar evolution code to compute magnetic effects due to the Spruit-Taylor mechanism and the MRI, separately and together, in a sample of massive star models. We find that the MRI can be active in the later stages of massive star evolution, leading to mixing effects that are not captured in models that neglect the MRI. The MRI and related magneto-rotational effects can move models of given ZAMS mass across "boundaries" from degenerate CO cores to degenerate O/Ne/Mg cores and from degenerate O/Ne/Mg cores to iron cores, thus affecting the final evolution and the physics of core collapse. The MRI acting alone can slow the rotation of the inner core in general agreement with the observed "initial" rotation rates of pulsars. The MRI analysis suggests that localized fields ~10^{12} G may exist at the boundary of the iron core. With both the ST and MRI mechanisms active in the 20 solar mass model, we find that the helium shell mixes entirely out into the envelope. Enhanced mixing could yield a population of yellow or even blue supergiant supernova progenitors that would not be standard SN IIP.

The “Christmas burst” GRB 101225A revisited

AbstractLong GRBs are related to the death of massive stars and reveal themselves through synchrotron emission from highly relativistic jets. The ‘Christmas Burst’ GRB 101225A was an exceptionally long GRB with a thermal afterglow, very different from the standard GRB. Initially, no spectroscopic redshift could be obtained and SED modeling yielded z=0.33. A plausible model was a He-NS star merger where the He-star had ejected part of its envelope in the common envelope phase during inspiral. The interaction between the jet and the previously ejected shell can explains the thermal emission. We obtained deep spectroscopy of the host galaxy which leads to a correction of the redshift to z=0.847. Despite the higher redshift, our model is still valid and theoretically better justified than the alternative suggestion of a blue supergiant progenitor proposed by Levan et al. (2014) for several “ultra-long” GRBs.