Detectability of neutrino-signal fluctuations induced by the hadron-quark phase transition in failing core-collapse supernovae

We introduce a systematic and quantitative methodology for establishing the presence of neutrino oscillatory signals due to the hadron-quark phase transition (PT) in failing core-collapse supernovae from the observed neutrino event rate in water- or ice-based neutrino detectors. The methodology uses a likelihood ratio in the frequency domain as a test-statistic; it is employed for quantitative analysis of neutrino signals without assuming the frequency, amplitude, starting time, and duration of the PT-induced oscillations present in the neutrino events and thus it is suitable for analyzing neutrino signals from a wide variety of numerical simulations. We test the validity of this method by using a core-collapse simulation of a 17 solar-mass star by Zha et al. [Astrophys. J. 911, 74 (2021) ]. Based on this model, we further report the presence of a PT-induced oscillations quantitatively for a core-collapse supernovae out to a distance of $\sim 10\mathrm{kpc}$, $\sim 5\mathrm{kpc}$ for IceCube and to a distance of $\sim 10\mathrm{kpc}$, $\sim 5\mathrm{kpc}$, and $\sim 1\mathrm{kpc}$ for a 0.4 Mt mass water Cherenkov detector. This methodology will aid the investigation of a future galactic supernova and the study of hadron-quark phase in the core of core-collapse supernovae.

Figure 1

Central density (top), effective neutrinosphere temperature of ${\overline{\nu }}_e$ (middle), and luminosity of ${\overline{\nu }}_e$ (bottom) as a function of time in CCSN simulations with STOS (solid red) and STOS-B145 (solid black) EoS. The zigzag feature in $T_{R_{\nu }}$ is due to the finite discretization of simulation grid.

Figure 2

The upper panel shows the central density oscillation spectrum in models with different compactness parameter ${\xi }_{2.2}$ as marked by the legends, while the lower panel shows the oscillatory neutrino luminosity spectrum in models with different compactness. The vertical dotted lines indicate the peak frequencies in the spectrum of density oscillation.

Figure 3

Neutrino event rate at Hyper-K (upper panels) and IceCube (lower panels), with (right) and without (left) the shot noise at SNe distance $D=10\mathrm{kpc}$.

Figure 4

Neutrino event rate (upper 2 panels) and its power spectrum (lower 2 panels) based on ZOS21-STOS-B145 at Hyper-K from 900 to 1050 ms after bounce (left) and from 1000 to 1050 ms after bounce (right), for distance $D=10\mathrm{kpc}$.

Figure 5

Probability density distribution of $\mathcal{L}$ in Hyper-K at 5 kpc. The PDF of $\mathcal{L}$ is estimated based on the ZOS21-STOS-B145 and ZOS21-STOS models, in both PT (Red) and no-PT (black dashed) scenario. The upper left, upper right, lower left, and lower right panel corresponds to the PDF of $\mathcal{L}s$ before the PT-induced oscillation, during the PT-induced oscillation transition, during the PT-induced oscillation and after the PT-induced oscillation. The starting time and the duration of analyzed time series are shown at the top of each panel.

Figure 6

Probability of detection at false alarm rate being 10%, at 1 (upper left), 5 (upper right), and 10 (lower) kpc in Hyper-K. The vertical and horizontal axis are duration and starting time of analyzed neutrino time series, respectively.

Figure 7

Probability of detection at false alarm rate being 10%, at 5 (left), and 10 (right) kpc in IceCube. The vertical and horizontal axis are duration and starting time of analyzed neutrino time series respectively.

Figure 8

Averaged frequencies ${\tilde{f}}_{\mathrm{PT}}$ (left) and amplitude $a$ (right) in various time series, at 10 kpc in Hyper-K (upper) and IceCube (lower). The vertical and horizontal axis are duration and starting time of analyzed neutrino time series respectively.

Figure 9

Averaged frequencies ${\tilde{f}}_{\mathrm{PT}}$ (left) and amplitude $a$ (right) in various time series, at 5 kpc in Hyper-K (upper) and IceCube (lower). The vertical and horizontal axis are duration and starting time of analyzed neutrino time series respectively.

Figure 10

Averaged frequencies ${\tilde{f}}_{\mathrm{PT}}$ (left) and amplitude $a$ (right) in various time series, at 1 kpc in Hyper-K. The vertical and horizontal axis are duration and starting time of analyzed neutrino time series, respectively.

Figure 11

Top: time evolution of the central density in eight selected nonrotating 1D models with progenitor masses varying from $15.02M_{\odot }$ to $15.79M_{\odot }$. Bottom: time evolution of the neutrino luminosity in 8 selected nonrotating 1D models with progenitor mass varying from $15.02M_{\odot }$ to $15.79M_{\odot }$. Note that in models with progenitor mass of $15.49M_{\odot }$ and $15.63M_{\odot }$, the PCS immediately collapse to black holes upon reaching the PT and the neutrino luminosities terminate at $\sim 3.1\mathrm{s}$ and $\sim 3.2\mathrm{s}$ after the first bounce.