Dynamically Driven D⁶ Double-Detonation Model

• The D⁶ model is a double-degenerate system where a helium shell detonation on the primary white dwarf dynamically triggers a secondary carbon–oxygen core detonation.
• Simulations reveal distinct ejecta asymmetries, specific nucleosynthetic yields, and a surviving hypervelocity white dwarf that serve as key observational signatures.
• Numerical resolution and ignition thresholds remain critical, with ongoing debates about helium shell mass requirements and the conditions for successful versus failed detonations.

The dynamically-driven double-degenerate double-detonation model, usually abbreviated $D^6$, is a double-degenerate, sub-$M_{\rm Ch}$ double-detonation channel for Type Ia supernovae in which dynamical interaction in a white-dwarf binary ignites a helium-shell detonation on the more massive primary, and that surface detonation then triggers a secondary carbon–oxygen core detonation. In the canonical formulation, only the primary explodes and the companion white dwarf survives as a hypervelocity remnant; in related realizations, the companion may also detonate. The model is also referred to as the “helium-ignited violent merger model” in part of the literature (Tanikawa et al., 2018, Tanikawa et al., 2019).

1. Definition and position among Type Ia progenitor models

In the $D^6$ model, both binary components are white dwarfs, typically with a carbon–oxygen primary and a helium-bearing outer layer that can be ignited during unstable mass transfer or merger. The defining sequence is a surface helium detonation followed by a core carbon–oxygen detonation in the primary, with the trigger supplied by dynamical binary interaction rather than by long-term secular accumulation from a non-degenerate donor (Tanikawa et al., 2018, Tanikawa et al., 2019).

This framing distinguishes $D^6$ from several adjacent Type Ia scenarios. It differs from classical single-degenerate double detonations, where a non-degenerate helium donor builds up a hydrostatic shell; from violent mergers or spiral-instability double-degenerate models, which typically destroy both white dwarfs; and from collisional double-degenerate models, in which the stars collide nearly head-on and no bound companion remains (Tanikawa et al., 2018). In its canonical form, $D^6$ also differs from broader “double-detonation in a double-degenerate system” language by requiring that the helium ignition be dynamically driven during the interaction phase and that the companion survive.

A recurrent theme in the literature is that the canonical $D^6$ picture uses a relatively thin helium layer, quoted as $\lesssim 0.01\,M_\odot$ in the “canonical” picture, whereas several numerical studies adopt thicker shells to guarantee ignition and resolve the detonation hydrodynamics (Tanikawa et al., 2018, Tanikawa et al., 2019). This gap between the “canonical” shell mass and the shell masses often used in simulations is central to the model’s numerical and observational interpretation.

2. Ignition physics and the primary double detonation

The physical trigger in $D^6$ is the interaction between the accretion stream and the primary’s helium-bearing surface. Local two-dimensional calculations of stream impact show that a directly impacting stream can produce a surface detonation for a range of stream parameters, that thinner helium shells ignite more promptly via impact, and that there are lower limits on ignition in both shell density and incoming stream speed (Rajavel et al., 2024). These results support stream ignition as a viable route to the first detonation in $D^6$.

Once the helium detonation is launched, the core ignition mechanism is shock convergence. In the three-dimensional explosion-phase calculation of a $1.0\,M_\odot$ primary with a $M_{\rm Ch}$0 He–C–O shell and a $M_{\rm Ch}$1 carbon–oxygen companion, the helium detonation sweeps around the shell, a shock separates from the helium detonation wave and propagates inward, and the shock converges in an off-center region of the core at $M_{\rm Ch}$2 s; by $M_{\rm Ch}$3 s the temperature at the convergence point becomes high enough to ignite carbon, after which a carbon detonation propagates through the core and destroys the primary (Tanikawa et al., 2018). The same study explicitly identified the mechanism as a He-shell detonation launching a converging shock that directly ignites the carbon–oxygen core off-center.

The robustness of this mechanism against rapid rotation has also been tested. Three-dimensional simulations of rotating white dwarfs found that when helium ignition is located far from the spinning axis the detonation fronts converge asynchronically at the antipodes of the igniting point, but the detonation of the carbon core still remains as the most probable outcome (García-Senz et al., 2018). This is directly relevant to merger-driven systems, for which rapid spin-up of the primary is plausible.

At the same time, ignition remains numerically delicate. A one-dimensional study of the helium envelope in dynamical accretion found that the chance of direct initiation of detonation is limited, that spontaneous detonation requires the primary envelope mass of $M_{\rm Ch}$4 solar mass, and that ignition as deflagration is instead far more likely, feasible for the lower envelope mass down to $M_{\rm Ch}$5 solar mass, with a possible subsequent detonation if a deflagration-to-detonation transition is realized (Iwata et al., 2022). This does not negate multidimensional stream-triggered ignition, but it makes clear that the ignition problem is not closed.

3. Outcome space and the fate of the secondary

The term $M_{\rm Ch}$6 is most strictly applied to the branch in which only the primary white dwarf detonates. However, the same dynamical configuration admits several nearby outcomes, depending on the companion type, its helium shell, and the orbital separation.

Outcome	Defining feature	Representative implications
Canonical $M_{\rm Ch}$7	Primary double-detonates; companion survives	Hypervelocity white dwarf remnant (Tanikawa et al., 2018)
TD/QD branches	Companion also detonates	“Triple” or “quadruple” detonation taxonomy (Tanikawa et al., 2019)
Failed $M_{\rm Ch}$8	Helium layer detonates but CO core does not	Faint nova-like transient, not luminous SN Ia (Roy et al., 2022)

A broader grid of double-degenerate double-detonation simulations found three regimes. In the $M_{\rm Ch}$9 regime, only the primary WD explodes and the companion survives. In the “triple-detonation” branch, the companion He white dwarf detonates after the primary double detonation. In the “quadruple-detonation” branch, a carbon–oxygen companion with a sufficiently thick helium shell undergoes its own double detonation. In that study, $D^6$0 outcomes occurred when the companion was too far, too low-mass, or had a helium shell too thin, $D^6$1, for the primary’s blast wave to ignite it (Tanikawa et al., 2019).

The uncertainty surrounding the secondary’s fate is illustrated by self-consistent merger simulations of a $D^6$2 primary and $D^6$3 secondary, both with $D^6$4 helium shells. In that system, the primary consistently detonates. A helium detonation around the secondary occurs, but the secondary carbon detonation fails in the self-consistent run; a restart with a carbon detonation ignited in the secondary by hand provides a second branch for comparison (Pakmor et al., 2022). The outer ejecta at $D^6$5 km s$D^6$6 are indistinguishable between the two branches, and light curves and spectra are very similar until $D^6$7 d after explosion, but the inner ejecta differ significantly and the decline rate of the bolometric light curve after maximum is slowed by $D^6$8 per cent when the secondary explodes (Pakmor et al., 2022).

Three-dimensional radiative transfer for the same pair of hydrodynamic models found that both the surviving-secondary and detonating-secondary realizations match properties of the peculiar 02es-like subclass, while a closer spectroscopic match to normal Type Ia supernovae can be obtained when the secondary detonates and the effects of helium detonation ash are minimised (Pollin et al., 2024). This suggests that the fate of the secondary primarily controls inner-ejecta asymmetry, angle dependence, and late-time diagnostics rather than the gross early-time appearance alone.

4. Ejecta structure, nucleosynthesis, and predicted observables

Successful $D^6$9 explosions synthesize a recognizable combination of iron-group elements, intermediate-mass elements, helium-detonation ash, and companion-related asymmetries. In the $D^6$0 three-dimensional simulation, the ejecta at $D^6$1 s contain $D^6$2 of $D^6$3, $D^6$4 of Si-group material from the CO detonation, $D^6$5 of heavy Si-group products from the He detonation dominated by Ca, and a companion-origin stream of $D^6$6 of carbon–oxygen stripped from the companion (Tanikawa et al., 2018). The same calculation predicts a surviving companion velocity of $D^6$7 km s$D^6$8, captured $D^6$9 of $D^6$0, captured He of $D^6$1, and a bulk ejecta velocity shift of $D^6$2 km s$D^6$3 due to the orbital motion of the exploding primary (Tanikawa et al., 2018).

The companion-origin stream is one of the most distinctive $D^6$4 signatures. In the broader model grid, the stripped mass is $D^6$5, and when the companion is a carbon–oxygen white dwarf with a helium shell $D^6$6 the stripped material contains carbon and oxygen and contributes to low-velocity ejecta components (Tanikawa et al., 2019). This is the basis for proposals that $D^6$7 explosions can be counterparts of sub-luminous SNe Ia and may contribute to low-velocity C seen in several SNe Ia (Tanikawa et al., 2019).

Late-time spectroscopy is sensitive to these asymmetries. Full non-local thermodynamic equilibrium nebular calculations showed that multidimensional structures significantly alter the overall ionisation balance, width and velocity of features, especially when the secondary detonates. Both the primary-only and two-explosion realizations produce most observed features from the optical to mid-infrared, but the current model realisations do not consistently reproduce all line shapes or relative strengths, and they yield prominent optical Ar III emission inconsistent with the data (Pollin et al., 7 Jul 2025). When the secondary detonates, including 3D effects improves the average agreement with observations, but weak optical Co III emission together with optical O I and near-infrared S I challenge its viability for normal Type Ia supernovae; the present comparisons tentatively favour detonation of only the primary white dwarf (Pollin et al., 7 Jul 2025).

An additional observational layer is provided by recent work on early emission from double detonations. Because $D^6$8 is a double-detonation channel, these predictions are directly relevant: a shock breakout flash, a stage of planar shock breakout cooling, and shock cooling emission from the thermal energy released by the collision of the CO detonation with the He detonation are expected. The quoted fiducial sequence is an initial flash dominated by the planar phase of $D^6$9, lasting $D^6$0 s in the soft X-rays, followed by $D^6$1–24 hrs of shock cooling at a luminosity of $D^6$2–$D^6$3 in the optical/UV (Piro et al., 18 Jul 2025).

5. Surviving white dwarfs, supernova remnants, and empirical diagnostics

A defining empirical consequence of canonical $D^6$4 is the surviving hypervelocity white dwarf. A Gaia DR2 search for such objects identified three candidates with total Galactocentric velocities between $D^6$5 and $D^6$6 km s$D^6$7, consistent with having previously been companion WDs in pre-SN Ia systems; one of the three has a past position within a faint, old supernova remnant, strengthening the interpretation as a hypervelocity runaway from a binary that underwent a Type Ia supernova (Shen et al., 2018). This is the most direct observational argument that at least some Type Ia supernovae arise from a $D^6$8-like channel.

The remnant phase preserves complementary signatures. A supernova-to-remnant calculation based on a $D^6$9 explosion showed that the first detonation produces an ejecta tail visible at early times, the second detonation leaves a central density peak in the ejecta that is visible at late times, and the supernova-remnant shell is off-centre at all times because of an initial velocity shift due to binary motion (Ferrand et al., 2022). The companion white dwarf produces a large conical shadow in the ejecta, visible in projection as a dark patch surrounded by a bright ring, and this feature is localized, long-lasting, and viewing-angle dependent (Ferrand et al., 2022). Such remnant-scale diagnostics are especially valuable because they probe the binary geometry directly rather than only the radiative outcome of the thermonuclear burning.

Radiative-transfer studies of the merger models reinforce this diagnostic program. The asymmetry in the width-luminosity relationship in the three-dimensional synthetic observables is comparable to previous double-detonation models, but the overall spread is increased when the secondary detonates (Pollin et al., 2024). This suggests that orientation dependence is not a secondary correction but part of the model’s observable identity.

6. Numerical status, controversies, and open problems

The $\lesssim 0.01\,M_\odot$0 literature is unusually explicit about numerical sensitivity. A two-dimensional resolution study of the double-detonation mechanism found three phases—external helium-rich detonation, core compressive heating, and a final core carbon burn—but only models with minimum resolutions of 4 km and better exhibit all three phases; particularly, core compression heating is only observed for higher resolutions, and finer spatial resolution alters the mixing of hot silicon at the shell–core interface and therefore the nucleosynthetic outcome (Rivas et al., 2022). This implies that global merger calculations, which are generally coarser than dedicated ignition calculations, may mischaracterize the mode of core ignition.

Ignition itself remains contested. The one-dimensional dynamical-accretion study already noted indicates that spontaneous helium detonation requires the primary envelope mass of $\lesssim 0.01\,M_\odot$1 solar mass, whereas lower envelope masses more readily ignite as deflagrations and may require a later deflagration-to-detonation transition (Iwata et al., 2022). By contrast, local two-dimensional stream-impact simulations show that directly impacting streams can produce a surface detonation for a range of stream parameters and that thinner helium shells ignite more promptly (Rajavel et al., 2024). This suggests that multidimensional compression, focusing, and stream geometry are not details but may be decisive.

A further complication is that not every helium ignition in a double-degenerate merger produces a successful Type Ia supernova. Three-dimensional hydrodynamical simulations of helium-ignited double-degenerate mergers found that the helium layer of the accreting primary white dwarf does undergo a detonation, while the underlying carbon–oxygen core does not, leading to an extremely rapid and faint nova-like transient instead of a luminous SN Ia event (Roy et al., 2022). The authors argued that this failed core detonation suggests that $\lesssim 0.01\,M_\odot$2 SNe Ia may be restricted to the most massive carbon–oxygen primary white dwarfs, and they highlighted the nucleosynthesis of the long-lived radioisotope $\lesssim 0.01\,M_\odot$3Ti during explosive helium burning as a hallmark of both successful and failed $\lesssim 0.01\,M_\odot$4 events (Roy et al., 2022).

Taken together, the current research picture is internally coherent but not final. $\lesssim 0.01\,M_\odot$5 is supported by successful three-dimensional primary detonations, by a physically motivated stream-impact ignition route, by hypervelocity white-dwarf candidates, and by a distinctive set of ejecta, nebular, and remnant asymmetries (Tanikawa et al., 2018, Rajavel et al., 2024, Shen et al., 2018). At the same time, the channel remains constrained by ignition thresholds, interface resolution, helium-shell ash, the uncertain fate of the secondary, and the existence of failed core detonations (Rivas et al., 2022, Iwata et al., 2022, Roy et al., 2022). A plausible implication is that “$\lesssim 0.01\,M_\odot$6” is best regarded not as a single explosion model but as a dynamical family of double-degenerate double detonations whose astrophysical relevance depends sensitively on primary mass, shell structure, companion properties, and three-dimensional flow geometry.