Formation of circumstellar material during double-white-dwarf mergers and the early excess emissions in Type Ia supernovae

TL;DR

The paper tackles the origin of early-time flux excess in some Type Ia supernovae by modeling circumstellar matter (CSM) formation during double white-dwarf mergers, with a focus on the violent merger (VM) channel. It develops a wind-driven, super-Eddington mass-transfer framework for DD binaries and derives a universal CSM density profile $\rho_{\rm CSM}(r) \propto r^{-3.5}$ with $D \simeq 10^{-14}$–$10^{-13}$ g cm$^{-3}$, yielding CSM masses of $\sim 0.01$–$0.03\,M_\odot$. Hydrodynamic and light-curve simulations using STELLA with W7 ejecta show that SN–CSM interaction produces early optical/UV light-curve excesses, peaking within a few days and reaching UV-bright values ($uvw2$) of $\sim -16$ to $-17$ mag, in qualitative agreement with 03fg/02es-like SNe. The work further discusses observational diagnostics (emission lines, X-ray signatures), CSM geometry, the He-ignited VM case, and the connection to hypervelocity stars, highlighting UV/X-ray observations as decisive tests and linking the progenitor channel to a broader set of phenomena.

Abstract

Early excess emission observed in Type Ia supernovae (SNe Ia) within $\sim1$ day of explosion provides a critical window into their progenitor systems. In the present study, we investigate formation of the circumstellar matter (CSM) in double white-dwarf (WD) mergers. We further study the interaction between the CSM and the SN ejecta. We first model the orbital evolution and super-Eddington mass transfer/ejection in the double WD systems. We then conduct hydrodynamical and light-curve (LC) simulations of the SN-CSM interaction, assuming a prompt SN Ia explosion in a context of the carbon-ignited violent merger (C-ignited VM). Our simulations show that at the moment of the merger, the binary system has the CSM distribution following $ρ_{\mathrm{CSM}}\simeq D(r/10^{14}\ \mathrm{cm})^{-3.5}\ (D\simeq 10^{-14}\text{--}10^{-13}\ \rm g\ cm^{-3})$. The simulated LCs reproduce the early flux excesses across optical to UV bands, as well as their color evolution, observed in the VM candidates, i.e., 03fg/02es-like SNe Ia. This supports that 03fg/02es-like objects originate from the VM explosions. We also discuss the case of the helium-ignited VM, which might be realized in some WD-WD mergers depending on the He content in the system. Focused here is the timing when the explosion is initiated, and we find that the explosion is initiated after the companion WD is, at least partially, tidally disrupted also in this case; we thus expect the formation of the CSM through the mass transfer phase also for the helium-ignited VM scenario.

Figures (8)

• Figure 1: The evolution of the orbital separations ($a$: left panels) and the mass-transfer rates ($\dot{M}_{\mathrm{D}}$ in unit of the Eddington's accretion limit: right panel). The solid and dotted lines are for $M_\mathrm{A}=1.1\ M_\mathrm{\odot}$ and $0.9\ M_\mathrm{\odot}$, respectively. Different line colors are used for different donor WD masses; $M_\mathrm{D}=1.0\ M_\mathrm{\odot}$ (blue), $0.8\ M_\mathrm{\odot}$ (green), and $0.6\ M_\mathrm{\odot}$ (purple). The Eddington accretion rate is typically $\sim10^{-5}\ M_{\odot}\ \mathrm{yr^{-1}}$ in any models.
• Figure 2: The transferred mass before the merger (i.e., when the separation reaches the tidal radius).
• Figure 3: The CSM mass from by the evolution toward the merger.
• Figure 4: The CSM density distribution for the same set of the models from Figure \ref{['fig:orbitalevolution']}. Also shown is the W7 ejecta model (brawn). The black solid line shows the slope of $r^{-3.5}$. The inset is an expanded view at $5\times10^{12}\text{--}10^{14}$ cm. The red points represent the shock-breakout radii, corresponding to the optical depth of 30 measured from the outside (i.e., the shock velocity is assumed to be $10000\ \mathrm{km\ s^{-1}}$). Thomson scattering in hydrogen-poor matter, $0.2\ \mathrm{cm^{2}\ g^{-1}}$, is used for the opacity. The CSM density structure for the steady-state mass loss is given by the grey lines (assuming a constant velocity of $v_{\mathrm{wind}}=4000\ \mathrm{km\ s^{-1}}$).
• Figure 5: The synthesized photometric evolutions for the same set of models from Figure \ref{['fig:orbitalevolution']}. Also shown is the no-CSM model (grey). Shown here are (A) the pseudo-bolometric LCs (3250-8900 Å), (B) the $g$-band LCs, (C) the $g-r$ color evolution, and (D) the $uvw2$-band LCs.