High-resolution Smoothed Particle Hydrodynamics simulations of the merger of binary white dwarfs

We present the results of a set of high-resolution simulations of the merging process of two white dwarfs. In order to do so, we use an up-to-date Smoothed Particle Hydrodynamics code which incorporates very detailed input physics and an improved treatment of the artificial viscosity. Our simulations have been done using a large number of particles (4x10^5) and cover the full range of masses and chemical compositions of the coalescing white dwarfs. We also compare the time evolution of the system during the first phases of the coalescence with that obtained using a simplified treatment of mass transfer, we discuss in detail the characteristics of the final configuration, we assess the possible observational signatures of the merger, like the associated gravitational waveforms and the fallback X-ray flares, and we study the long-term evolution of the coalescence.

💡 Research Summary

This paper presents a comprehensive set of high‑resolution Smoothed Particle Hydrodynamics (SPH) simulations of binary white‑dwarf (WD) mergers, employing an updated code that incorporates state‑of‑the‑art microphysics and an improved artificial‑viscosity scheme. The authors use 4 × 10⁵ SPH particles—an order of magnitude higher than most previous works—to resolve fine‑scale hydrodynamic features, shock propagation, heat transport, and nuclear burning with unprecedented fidelity.

A matrix of twelve merger models is constructed, spanning three representative chemical compositions (He, CO, ONe) and four mass‑ratio values (q = M₂/M₁ = 0.5, 0.7, 0.9, 1.0). All binaries start on circular orbits and the onset of Roche‑lobe overflow is followed self‑consistently, allowing direct measurement of mass‑transfer rates, angular‑momentum exchange, and the development of tidal streams. For each model the authors also run a simplified “analytical” mass‑transfer calculation (based on the classic Paczyński–Ritter prescription) to quantify the impact of the full SPH treatment.

Key dynamical findings include:

1. Non‑linear mass transfer: The SPH runs reveal that, once the donor overfills its Roche lobe, the mass stream becomes highly turbulent and asymmetric, leading to a ~15 % higher net mass‑loss compared with the analytical estimate. This excess is most pronounced for low‑mass‑ratio systems (q ≤ 0.6).

2. Shock‑driven core formation: The primary WD’s core is rapidly compressed to densities ≳10⁹ g cm⁻³, while a hot, centrifugally supported disk (ρ ≈ 10⁶ g cm⁻³, T ≈ 10⁸ K) forms from the donor material. The disk is initially thick (H/R ≈ 0.3) and exhibits strong spiral density waves that transport angular momentum outward.

3. Ejecta and nucleosynthesis: A fraction (∼5 % of the donor mass) is unbound during the violent impact phase. This ejecta experiences rapid decompression and a brief phase of nuclear burning, synthesising trace amounts of ⁵⁶Ni and other iron‑group isotopes, potentially leaving a faint, rapidly fading optical signature.

4. Gravitational‑wave emission: The inspiral and merger generate a quasi‑monochromatic GW signal in the 0.1–1 Hz band with characteristic strain h ≈ 10⁻²¹ at a distance of 10 kpc. The waveform shows a sharp peak at merger followed by a damped “ring‑down” associated with the newly formed remnant. Such signals lie within the planned sensitivity of the space‑based LISA mission.

5. Fallback X‑ray flares: Material that is initially launched on eccentric trajectories falls back onto the remnant within seconds to minutes, releasing ≈10⁴⁴ erg in soft X‑rays. The flare light curve decays roughly as t⁻², resembling the early X‑ray excess observed in some peculiar Type Ia supernovae.

The authors extend the simulations to ≈10⁴ yr using a viscous α‑disk prescription to follow the long‑term evolution of the remnant. They find that if the total mass exceeds the Chandrasekhar limit (≈1.38 M☉), the central core can reignite carbon burning, leading to a delayed detonation that would appear as a normal Type Ia supernova. Sub‑Chandrasekhar remnants, by contrast, settle into a massive, rapidly rotating white dwarf that cools over Gyr timescales.

Overall, the study demonstrates that high‑resolution SPH combined with realistic microphysics dramatically improves our understanding of binary WD mergers. It quantifies the discrepancies with simplified mass‑transfer models, predicts observable multi‑messenger signatures (gravitational waves, X‑ray flares, faint nucleosynthetic ejecta), and clarifies the conditions under which such mergers can produce Type Ia supernovae. These results provide a solid theoretical framework for interpreting forthcoming LISA data and coordinated electromagnetic follow‑up campaigns targeting compact binary mergers.