Neutrino transport in black hole-neutron star binaries: Neutrino emission and dynamical mass ejection

We study the merger of black hole-neutron star binaries by fully general-relativistic neutrino-radiation-hydrodynamics simulations throughout the coalescence, particularly focusing on the role of neutrino irradiation in dynamical mass ejection. Neutrino transport is incorporated by an approximate transfer scheme based on the truncated moment formalism. While we fix the mass ratio of the black hole to the neutron star to be 4 and the dimensionless spin parameter of the black hole to be 0.75, the equations of state for finite-temperature neutron-star matter are varied. The hot accretion disk formed after tidal disruption of the neutron star emits a copious amount of neutrinos with the peak total luminosity $\sim 1–3\times {10}^{53}\mathrm{erg}{\mathrm{s}}^{-1}$ via thermal pair production and subsequent electron/positron captures on free nucleons. Nevertheless, the neutrino irradiation does not modify significantly the electron fraction of the dynamical ejecta from the neutrinoless $\beta$-equilibrium value at zero temperature of initial neutron stars. The mass of the wind component driven by neutrinos from the remnant disk is negligible compared to the very neutron-rich dynamical component, throughout our simulations performed until a few tens milliseconds after the onset of merger, for the models considered in this study. These facts suggest that the ejecta from black hole-neutron star binaries are very neutron rich and are expected to accommodate strong $r$-process nucleosynthesis, unless magnetic or viscous processes contribute substantially to the mass ejection from the disk. We also find that the peak neutrino luminosity does not necessarily increase as the disk mass increases, because tidal disruption of a compact neutron star can result in a remnant disk with a small mass but high temperature.

Figure 1

Rest-mass density profiles in the central region at 10 ms after the onset of merger, which approximately corresponds to the peak of neutrino luminosity. The black circles show the apparent horizons. The velocity vector $v^i\equiv u^i/u^t$ is overplotted.

Figure 2

The same as Fig. 1 but for the temperature. We also show isodensity contours for $\rho =10^8$ (yellow dotted), $10^{10}$ (orange dashed), and $10^{12}\mathrm{g}{\mathrm{cm}}^{-3}$ (light blue solid).

Figure 3

The same as Fig. 1 but for the electron fraction. We also show isodensity contours for $\rho =10^8$ (yellow dotted), $10^{10}$ (orange dashed), and $10^{12}\mathrm{g}{\mathrm{cm}}^{-3}$ (light blue solid). The mass of material with a high electron fraction in the polar region is very small.

Figure 4

Neutrino luminosity curves for SFHo (top), DD2 (middle), and TM1 (bottom). The purple-dashed, green-solid, and cyan-dotted curves correspond to the electron neutrino, electron antineutrino, and one of the heavy-lepton neutrinos, respectively. The luminosity curves for the heavy-lepton neutrinos are multiplied by a factor of 10 to make the plot visible.

Figure 5

Time evolution of the mass remaining outside the apparent horizon for SFHo (purple dashed), DD2 (green solid), and TM1 (cyan dotted).

Figure 6

Time evolution of the maximum temperature in the accretion disk for SFHo (purple dashed), DD2 (green solid), and TM1 (cyan dotted).

Figure 7

Time evolution of the maximum rest-mass density in the accretion disk for SFHo (purple dashed), DD2 (green solid), and TM1 (cyan dotted).

Figure 8

Time evolution of the average neutrino energy for ${\nu }_e$ (top), ${\overline{\nu }}_e$ (middle), and ${\nu }_x$ (bottom). The purple-dashed, green-solid, and cyan-dotted curves correspond to SFHo, DD2, and TM1, respectively. The average energy of ${\overline{\nu }}_e$ and ${\nu }_x$ approaches 30 MeV at $t-t_{\text{merge}}\approx 3–4\mathrm{ms}$, whereas we restrict the range in this figure to focus on long-term behavior.

Figure 9

Rest-mass density (top), temperature (middle), and electron fraction (bottom) profiles on the equatorial plane at 3 ms after the onset of merger, when the mass and electron fraction of the ejecta approximately settle to asymptotic values. Note the different spatial scale compared to Figs. 1, 2, and 3. Black (top) and white (middle and bottom) curves indicate the unbound component identified by the conditions $u_t<-1$ (shown only for $\rho \ge 10^{6.5}\mathrm{g}{\mathrm{cm}}^{-3}$ to match the top row), which are indistinguishable from $h{u}_t<-1$ on panels presented here [11]. The velocity vector $v^i\equiv u^i/u^t$ is overplotted on the rest-mass density profiles. We also show isodensity contours for $\rho =10^7$ (yellow dotted), $10^9$ (orange dashed), and $10^{11}\mathrm{g}{\mathrm{cm}}^{-3}$ (light blue solid) on the temperature and electron-fraction profiles.

Figure 10

Time evolution of the ejecta mass identified by the condition $u_t<-1$ for SFHo (purple dashed), DD2 (green solid), and TM1 (cyan dotted).

Figure 11

Time evolution of the ejecta velocity estimated in the same manner as described in Refs. [11, 15] for SFHo (purple dashed), DD2 (green solid), and TM1 (cyan dotted). The values before $t=t_{\text{merge}}$ are numerical artifacts associated with the small ejecta mass. The values right after the onset of merger are $\approx 0.7c$, but we do not show such transitional values.

Figure 12

Time evolution of the mass-averaged electron fraction of the ejecta for SFHo (purple dashed), DD2 (green solid), and TM1 (cyan dotted).

Figure 13

Distribution of the electron fraction in the unbound material at $\approx 10\mathrm{ms}$ after the onset of merger. The left axis shows the fraction of material in the bin of width $\mathrm{\Delta }{Y}_e=0.01$, and the right axis shows the differential distribution normalized by the total ejecta mass. The latter is independent of the bin width.

Figure 14

Comparison of results for the DD2 model between high ($\mathrm{\Delta }x=270\mathrm{m}$, solid curve) and low ($\mathrm{\Delta }x=400\mathrm{m}$, dashed curve) resolutions. The former is the same as those have been shown in other figures. The top-left, top-right, bottom-left, and bottom-right panels show the neutrino luminosity for all the flavors, the ejecta mass, the ejecta velocity, and the averaged electron fraction of the ejecta, respectively. The luminosity curves for the heavy-lepton neutrinos are multiplied by a factor of 10 to make the plot visible in the top-left panel. The velocity right after the onset of merger is out of the vertical range and is not shown in the bottom-left panel as described in the caption of Fig. 11.

Figure 15

Neutrino luminosity curves of all the flavors for DD2. The color is the same as Fig. 4. The solid and dashed curves are results for simulations with and without neutrino absorption, respectively, and thus the former is the same as the middle panel of Fig. 4. The luminosity curves for the heavy-lepton neutrinos are multiplied by a factor of 10 to make the plot visible.

Figure 16

Time evolution of the mass (left) and averaged electron fraction (right) of ejecta for the DD2 model with (solid) and without (dashed) neutrino absorption.