The role of the secondary white dwarf in a double-degenerate double-detonation explosion, in the supernova remnant phase

TL;DR

Using results from Pakmor et al. 2022, the authors extend double-degenerate double-detonation SN models into the SNR phase up to 1500 years to predict observable imprints. They compare OneExp (primary detonates alone) and TwoExp (secondary also detonates) scenarios, revealing a long-lived conical shadow from the companion and, in the nested case, an inner ejecta–reverse shock interaction that enhances X-ray emission from inner layers. The study shows that TwoExp yields distinct spatial abundance patterns—unburnt C at shadow edges and Fe-enriched inner regions—detectable by spatially-resolved X-ray spectroscopy with current and upcoming observatories. The results provide a path to distinguish double-detonation progenitors and inform SN modeling and SNR interpretation.

Abstract

Type Ia supernovae (SNe) are believed to be thermonuclear explosions of white dwarf (WD) stars, but their progenitor systems and explosion mechanisms are still unclear. Here we focus on double degenerate systems, where two WDs are interacting, and on the double detonation mechanism, where a detonation of a helium shell triggers a detonation of the carbon-oxygen core of the primary WD. We take the results from three-dimensional SN simulations of Pakmor et al 2022 (arXiv:2203.14990) and carry them into the supernova remnant (SNR) phase, until 1500 yr after the explosion. We reveal signatures of the SN imprinted in the SNR morphology. We confirm the impact of a companion on the SNR: its presence induces a conical shadow in the ejecta, that is long lived. Its intersection with the shocked shell is visible in projection as a ring, an ellipse, or a bar, depending on the orientation. New, we test the case of a nested explosion model, in which the explosion of the primary induces the secondary to also explode. As the explosion of the secondary WD is weaker only the primary outer ejecta interact with the ambient medium and form the main SNR shell. The secondary inner ejecta collide with the reverse shock, which enhances the density and thus the X-ray emissivity. The composition at the points of impact is peculiar, since what is revealed are the outer layers from the inner ejecta. This effect can be probed with spatially-resolved X-ray spectroscopy of young SNRs.

Figures (11)

• Figure 1: Slices of the mass density at the end of the SN simulation (150 s for OneExp, 133 s for TwoExp), in the mid-plane along the $z$ axis. We have indicated the ejecta from the primary and from the secondary, as well as the main interaction regions: the angular sector (cyan, a cone in 3D) is the shadow in the primary ejecta from the presence of the secondary WD, the ring (purple, a shell in 3D) is the double-shock structure formed by the collision of the secondary ejecta into the primary ejecta. The curved segment inside the cone roughly indicates the spiraling motion of the secondary WD. A Sketchfab-based 3D interactive version of the OneExp and TwoExp SNe is available online.Sketchfab This version shows a set of isocontours of the mass density, rendered as semitransparent surfaces. Contours at 0.004%, 0.04%, 0.4%, 2.3%, 8%, 14%, 20% of the max were chosen to highlight the extent of the primary ejecta including the shadow and (for TwoExp) the interaction between the primary and secondary ejecta.
• Figure 2: Illustration of the making of the comparison maps. The quantity displayed is the mass density, as a slice through the centre of the box, along three different axes, at the age of 500 yr (see more details in section \ref{['sec:res_1']} and the time evolution in Figure \ref{['fig:maps_cut_OneTwoExp']}). The first two rows show data for OneExp and TwoExp respectively, while the last row shows the composite image that combines the two cases. Data in the first two rows are colourized using a sequential, grayscale colour map. The physical scale is linear but the colour bar has been stretched, by a power 0.33, for better visibility. The last row is a RGB composite of the first two rows, with the OneExp data used as the red channel and the TwoExp data used as the blue channel (the green channel is left at zero), each channel properly normalized in the range $[0,1]$. Where data overlap, the red and blue hues mix into tones of purples. This allows one to see at-a-glance the parts of the SNR that are specific to one case, or that are common to both. (Note that the boundary conditions are spherical, values are not defined in the corners of the box, that are rendered black.)
• Figure 3: Slices of the mass density at several times: 100 yr, 500 yr, 1000 yr after the explosion, along three different axes (principal axes $x$, $y$, $z$, of the simulation box). Each map compares the OneExp and TwoExp cases, with data from OneExp making the red channel and data from TwoExp making the blue channel (purple indicates overlap). Data ranges are normalized for each case and at each time, for each channel a power 0.33 function is applied. The maximum density decreases over time, from 100 yr to 1000 yr by a factor of about 125 for OneExp and 70 for TwoExp. Each case uses its own comoving grid, for the chosen ambient density, the box size is 5.47 pc $|$ 5.41 pc at 100 yr, 14.5 pc $|$ 14.9 pc at 500 yr, 19.4 pc $|$ 20.7 pc at 1000 yr for OneExp$|$TwoExp respectively. An animated version of this figure is available in the online journal, showing the evolution from 1 yr to 1500 yr in steps of 1 yr.
• Figure 4: Morphology of the contact discontinuity. Maps on the left are spherical (Mollweide) projections of the radial variations of the location of the edge of the ejecta. For all times the map is centred on the dipole component at the initial time. Spectra on the right result from an expansion in spherical harmonics of these variations. At angular wavenumber $\ell$, the typical angular scale probed is $\pi/\ell$, and the power $C_{\ell}$ plotted is normalized in such a way that each grayed bin is the contribution of wavenumber $\ell$ to the total variance of the radial fluctuations. One time is shown: 500 yr; maps and spectra evolve quickly in the first few hundred years, and slowly after that time. The two cases OneExp and TwoExp are compared at the top and bottom. An animated version of this figure is available in the online journal showing the evolution from 1 yr to 1500 yr in steps of 1 yr. Note that data ranges are adjusted for each case and at each time.
• Figure 5: Evolution of the angular power as a function of time, for the three waves: forward shock in red, contact discontinuity in green, and reverse shock in blue. The power is integrated over all angular scales (sum of the gray area on the power spectra), the quantity plotted is its root mean square (RMS). Time is indicated in years and in the characteristic timescale $t_\mathrm{ch} = {r_\mathrm{ch}}/{u_\mathrm{ch}}$ where $r_\mathrm{ch} = \left((3 M_\mathrm{ej})/(4 \pi \rho_\mathrm{ISM})\right)^{1/3}$ and $u_\mathrm{ch} = \left((2E_\mathrm{SN})/(M_\mathrm{ej})\right)^{1/2}$. The two cases OneExp and TwoExp are compared on the left and right.