CORDIS - EU research results

A comprehensive and Multi-facetted Study of Stellar Explosions and Eruptions using Radiation-Hydrodynamics and Time-Dependent Radiative Transfer Techniques

Final Report Summary - STELLAR EXPLOSIONS (A comprehensive and Multi-facetted Study of Stellar Explosions and Eruptions using Radiation-Hydrodynamics and Time-Dependent Radiative Transfer Techniques.)

The first goal of the proposal was to develop numerical tools suitable for the modeling of stellar explosions. The nature of the problem requires both the treatment of the explosion mechanism with radiation hydrodynamics, as well as the treatment of energy transport with non-Local-Thermodynamic-Equilibrium (non-LTE) time-dependent radiative transfer. In addition, full physical consistency can only be achieved if such simulations are based on physically-consistent models of the progenitor star.

The second goal of the proposal was to apply such tools to a wide range of explosive events, including stellar eruptions, core-collapse supernovae (SNe), pair-instability SNe, as well as the thermonuclear runaway in white dwarfs leading to a Type Ia SN.

Although the proposal appeared very ambitious when submitted in 2008, our results and publication record testify that all these key goals have been reached. On the modeling front, we have developed the radiation-hydrodynamics code V1D, in collaboration with Eli Livne and Roni Waldman (Racah Institute of Jerusalem), to treat core-collapse SNe and stellar eruptions, including grey radiation transport and explosive nucleosynthesis. These simulations are started from physically-consistent progenitor models produced with the public stellar-evolution code MESA-Star. We have developed considerable expertise with that code to generate a variety of initial conditions and be fully independent in our work. But the real achievement is the development and testing of CMFGEN, in collaboration with D. John Hillier (Pittsburgh University), to include all the relevant processes for radiative transfer in homologously expanding SN ejecta. We had included non-LTE and time-dependence. We now also treat non-thermal processes, non-local energy deposition associated with radioactive decay, and we have done considerable work in testing the dependence of SN radiation properties on non-LTE processes included and the atomic data employed.

The main deliverable for the project is the unique association of three codes, i.e. MESA, V1D, and CMFGEN for an end-to-end modeling of massive star explosions from main sequence to SN. While there exists other stellar evolution codes (e.g. KEPLER; Weaver, Zimmerman, & Woosley 1978, ApJ, 225, 102), other radiation hydrodynamics codes (e.g. STELLA; Blinnikov et al. 1998, ApJ, 496, 454), and other radiative-transfer codes (e.g. SEDONA; Kasen, Thomas, & Nugent, 2006, ApJ, 651, 366), the combination of MESA, V1D, and CMFGEN offers the ability to combine single as well as binary massive star models; explosion models with detailed nucleosynthesis; state-of-the-art non-LTE time-dependent radiative-transfer models for the light curve and spectra of Type II/Ib/Ic SNe. The assumption of spherical symmetry in CMFGEN prevents the direct investigation of full 3D explosions, but it allows detailed work on the non-LTE aspect of radiative transfer, which other codes (in particular multi-D codes) cannot treat.

Type Ia SNe arise from the thermonuclear disruption of a white dwarf but it is unclear today whether a single or two degenerate stars are involved in the process. Furthermore, the nuclear burning represents a formidable challenge since it occurs on microscopic scale but influences the dynamics within the star on scales many orders of magnitude larger. Consequently, the modeling of the combustion is parametrized to mimic a deflagration, a detonation, or a combination of both. To study SNe Ia, we needed to connect to a SN Ia combustion expert and we thus developed a new collaboration with Alexei Khokhlov (University of Chicago), who provides us with physically-consistent models for different types of SN Ia ejecta. Such ejecta are then modeled with CMFGEN to generate light curves and spectra.

Additional efforts were also carried out to extend the polarization studies of Dessart & Hillier (2011). The code, which used to work for a single line and an overlapping continuum, now works in a "blanketed" mode and can thus produce a full polarized spectrum (including any number of lines) for any SN type (the code still assumes polarization stems exclusively from electron scattering, which holds best for Type II SNe, but less so for SNe Ia/Ib/Ic). This improvement is key for a better confrontation to observations, which we obtain from the ESO/Very-Large-Telescope through a collaboration (Dessart/Pignata/Leonard/Hillier).

Finally, to investigate strongly dynamical events in multi-D, I have implemented "supernova" modules in the code HERACLES of Edouard Audit (CEA/Saclay), including a suitable equation of state and frequency-dependent opacities. The first application of the code is to perform 1-D but also multi-D simulations of interacting SNe, to complement but also improve the previous work of Dessart et al. (2009).

The results from these efforts are numerous and concern all types of SN explosions. We studied Type II SNe to investigate the dependency of their radiation properties on the progenitor evolution and explosion energy (Dessart et al. 2013, MNRAS, 433, 1745). We investigated the nature of SNe Ib and Ic (Dessart et al. 2012, MNRAS, 424, 2139), in particular through the influence of non-thermal processes (Li, Hillier & Dessart, 2012, MNRAS, 426, 1671).

We published a technical paper on the non-LTE time-dependent radiative-transfer code CMFGEN (Hillier & Dessart 2012, MNRAS, 424, 252). We investigated the circumstances surrounding black-hole formation in GRB progenitor models (Dessart, O'Connor, & Ott, 2012, ApJ, 754, 76). We carried out an extensive study on the origin of super-luminous SNe. This included a critical assessment of the difference between SNe powered by radioactive decay and by magnetar power (Dessart et al. 2012, MNRAS, 426, 76), and one realization for super-luminous SNe resulting from pair production (Dessart et al. 2013, MNRAS, 428, 3227). Finally, we recently published a series of four studies on SNe Ia (Blondin et al. 2013, MNRAS, 433, 1745; Dessart et al., 2013, arXiv:1308.6352; Dessart et al. 2013, arXiv:1310.7747; Dessart et al. arXiv:1310.7750). The code development and application to observations represent a great achievement, which more than fulfill what we proposed in the original proposal. They provide key insights into the properties of SN ejecta properties, the explosion mechanism, and the progenitor stars, for a large part because of the physical consistency of the approach and the unparralel level of sophistication of the non-LTE time-dependent radiative transfer carried out in CMFGEN.

Being public and well tested, CMFGEN represents a unique tool that will benefit the SN community at large in the years to come.