The collision and merger products of stars do not look alike: A magnetohydrodynamics comparison

TL;DR

The paper addresses how collisions and mergers of massive main-sequence stars produce differently structured remnants. By performing 14 high-resolution 3D MHD simulations with AREPO, informed by 1D MESA stellar models, it isolates differences in mixing, rotation, internal structure, and magnetic fields between collisions (8 cases) and mergers (6 cases) for masses in the $5$–$10 \,M_igodot$ range. Key findings include substantial magnetic-field amplification in both channels, with mergers showing the strongest, large-scale ordered fields and magnetized bipolar outflows absent in collisions, as well as higher core hydrogen fractions in some merger products. These distinctions imply potentially different evolutionary paths and observable signatures, motivating follow-up work to track the long-term evolution of collision and merger remnants using stellar evolution codes.

Abstract

A significant fraction of stars experience close interactions, including collisions resulting from gravitational encounters and mergers within close binary systems. These processes can produce more massive stars that may give rise to relatively rare objects such as blue stragglers. Distinguishing the outcomes of collisions and mergers is challenging yet essential for interpreting observations. This study utilizes the magnetohydrodynamics code AREPO to simulate collisions and mergers of $5$ to $10 \,\mathrm{M}_{\odot}$ main-sequence stars, systematically comparing the properties of the resulting products. Both collisions and mergers yield more massive, strongly magnetized, rapidly and differentially rotating stars with cores enriched in hydrogen, but notable quantitative differences emerge. Merger products exhibit core hydrogen fractions up to $10\%$ higher than those of collision products. In both scenarios, turbulent mixing amplifies magnetic field energies by $9$ to $12$ orders of magnitude. However, magnetic fields in small-impact-parameter collision products display small-scale reversals that may dissipate over time, whereas merger products and large-impact-parameter collision products develop large-scale ordered, potentially long-lived magnetic fields. Additionally, only merger products display magnetically driven, bipolar outflows with radial velocities exceeding $300$ to $400 \,\mathrm{km}\,\mathrm{s}^{-1}$. These distinctions may result in divergent long-term evolutionary outcomes, which warrant further investigation in future studies.

Figures (7)

• Figure 1: Initial density and temperature profiles of $10 \,\mathrm{Myr}$-old MS stars of different masses, corresponding to Table \ref{['tab:star_models']}. The thinner dashed and the thicker solid lines represent the 1D profiles from MESA and 3D shell-averaged and relaxed profiles from AREPO, respectively. The AREPO profiles diverge from their MESA counterparts at densities of $\lesssim 10^{-4} \,\mathrm{g}\,\mathrm{cm}^{3}\,\mathrm{s}^{-1}$ and temperatures of $\lesssim 10^{6.25} \,\mathrm{K}$.
• Figure 2: Mid-plane snapshots of a stellar collision between a $10 \,\mathrm{M}_\odot$ star and a $5 \,\mathrm{M}_\odot$ star when $b = 0.25$, $10 \,\mathrm{Myr}$ into their MS lifetimes. The three rows illustrate slices, in the collision plane, of densities $\rho$ (top), passive scalars $K$ (center), and magnetic field magnitudes $B$ (bottom). The four columns illustrate these quantities at the times $t$ before collision (leftmost), just before collision (center-left), just after collision (center-right), and after $\sim 60\,t_{\mathrm{dyn},r_{\mathrm{tot}}}$ following collision (rightmost). The core of the collision product is dominated by the initially less massive star, while the outer layers are composed of material from both stars in varying fractions. Furthermore, the amplification of the magnetic field is clearly visible across the panels.
• Figure 3: Mid-plane snapshots, similar to Figure \ref{['fig:coll_collage_evo']}, of a stellar merger between a $10 \,\mathrm{M}_\odot$ star and a $5 \,\mathrm{M}_\odot$ star, $10 \,\mathrm{Myr}$ into their MS lifetimes. The merger product has a central region dominated by the initially less massive star, an outer region dominated by the initially more massive star, and a transition region composed of material from both stars (see Figure \ref{['fig:coll_collage_evo']} for comparison with the collision product). Moreover, the magnetic field within the merger product is notably higher.
• Figure 4: Mass profiles of $^{1}\mathrm{H}$ fractions $X$ of collision (left) and merger (right) products for different masses. The dashed and solid lines in the collision plot represent $b = 0.50$ and $b = 0.25$, respectively. The shaded regions' boundaries represent the 15th and 85th percentiles. The resulting stars follow similar chemical profiles of $X$. As a comparison, Table \ref{['tab:star_models']} shows the core hydrogen fractions of the initial stars.
• Figure 5: Mid-plane radial profiles of mid-plane angular frequency ratios $\omega/\omega_{\mathrm{crit}}$ profiles of collision (left) and merger (right) products for different masses. Merger products rotate faster than collision products and are closer to break-up velocity.