Shunsaku Horiuchi Virginia Tech - - PowerPoint PPT Presentation

ニュートリノで解き明かす 大質量星の重力崩壊

Image credit: NASA/ESA

Shunsaku Horiuchi Virginia Tech

Outline

Opportunities with Galactic supernovae: transient (a selection) Beyond the Milky Way: diffuse Multi-messenger connections Concluding remarks

Shunsaku Horiuchi (Virginia Tech) 2

Collapse of massive stars

Core collapse (implosion)

R: 8000 km à ~20 km r: ~109 g cm-3 à ~1014 g cm-3 T: ~1010 K à ~30 MeV

Explosion

Adapted from slides by G. Raffelt

Massive (>8Msun) star structure

Fe

Si O C He H

Shunsaku Horiuchi (Virginia Tech) 3

The explosion mechanism

VS

Mass accretion n heating Explosion mechanism The bounce shock stalls – and must be energetically revived. Neutrino mechanism The neutrinos achieve this:

Mass accretion rate Neutrino luminosity Critical curve Explosions Explosion failure and black hole formation

The prevailing mechanism is turbulence- enhanced energy deposition

• 1D: fails
• 2D: many successful explosions
• 3D: success, but fewer simulations

Many groups: Australia, Germany, Japan, US

Shunsaku Horiuchi (Virginia Tech) 4

Neutrino heating mechanisms

For mass < 10 Msun For mass < 15 Msun For mass < 40 Msun Readily explodes by neutrino heating, even in spherical symmetry. Does not explode in spherical symmetry, but does explode in axisymmetric and in full 3D via convection

9.6 Msun, Melson et al 2015,

Kitaura et al 2006

11.2 Msun, Takiwaki et al 2014 20 Msun, Melson et al 2015

Does not explode in spherical symmetry, but does explode in axisymmetric and in full 3D with SASI Can neutrinos reveal the progenitor Mdot? Can neutrinos reveal the explosion mechanism?

Shunsaku Horiuchi (Virginia Tech) 5

Neutrinos: highlights

Three main phases of supernova neutrino emission:

Burst Accretion Cooling Core collapse timing Explosion mechanism Proto-neutron star info Supernova triangulation Progenitor information Exotic energy loss Standard candle Supernova pointing Nucleosynthesis Oscillations may be simple Collective oscillations Shock effects Neutrino mass ordering … Nuclear physics … …

Shunsaku Horiuchi (Virginia Tech) 6

Progenitor dependence

VS !

Mass accretion Neutrino heating

Higher x & Mdot L

• w

e r x & M d

• t

Compactness: A simple way to capture the density structure Core density profile: Progenitor core density profile critical for collapse evolution

• Higher x à higher Mdot
• Lower x à lower Mdot

O’Connor & Ott (2011)

ξM = M/M R(Mbary = M)/1000 km

• t

Shunsaku Horiuchi (Virginia Tech) 7

Accretion phase: progenitor information

Progenitor dependence: The neutrino emission reflects the compactness of the progenitor

Horiuchi et al (2017) O’Connor & Ott (2013)

Spherical simulations èèè Axisymmetric simulations

(d=10 kpc)

Shunsaku Horiuchi (Virginia Tech) 8

Accretion phase: SASI-driven explosion?

Tamborra et al (2013, 2014), based on Hanke et al (2013) see also Lund et al (2010, 2012)

Multiple SASI episodes + convection Single SASI episode + convection No SASI episode,

• nly convection

Sloshing spiral

Signatures:

• SASI’s time variations (~10-20 ms)

get imprinted on the neutrino luminosity and energy.

• Can be measured with large

volume detectors

(d=10 kpc)

Shunsaku Horiuchi (Virginia Tech) 9

Cooling phase

Late-time signal Detectable up to O(100) seconds post bounce Depends sensitively on the proto-neutron star mass

1.35Msun 2.05Msun 1 . 2 M s u n

Suwa et al (2019)

(d=10 kpc)

Shunsaku Horiuchi (Virginia Tech) 10

Did it collapse to a black hole?

High event rates predicted

O’Connor (2015) Liebendoerfer et al (2004)

Moment of black hole formation Neutrinos light curve can reveal the moment of black hole formation

BH case NS case Termination (but not to 0)

Shunsaku Horiuchi (Virginia Tech) 11

Distance, strategy, physics!

Nn >> 1 : BURST SN rate ~ 0.01 /yr Nn ~ 1 : MINI-BURST SN rate ~ 0.5 /yr Nn << 1 : DIFFUSE SN rate ~ 108 /yr

~kpc ~Mpc ~Gpc Galactic burst Mini-bursts Diffuse signal Physics reach Multi-messenger astronomy, explosion mechanism, neutrino physics supernova variety Average emission, multi-populations (e.g., black holes)

Adapted from Beacom (2012)

reviews by Beacom (2010), Lunardini (2010)

Shunsaku Horiuchi (Virginia Tech) 12

Landscape of supernova diversity

Systematic studies:

• Based on “neutrino engine” 1D simulation suite
• Characterizing in ZAMS mass looks incomplete

Janka 2017; based on Ertl et al (2016); see also Ugliano et al (2012), Sukhbold et al (2016), Pejcha & Thompson (2015), Mueller et al (2016)

Shunsaku Horiuchi (Virginia Tech) 13

Making sense of the landscape

Failure should correlate with high compactness Failed explosions correspond to stars with large compactness

Gray: failed explosions Red: explosions

• BH formation for x2.5 > 0.25
• Explosions for x2.5 < 0.15
• Mixture in between
• 1 compactness successful in up to ~88% of stars
• 2 parameters up to ~97% of stars

Erlt et al (2015), Pejcha & Thompson (2015)

ξM = M/M R(Mbary = M)/1000 km

• t

Compactness has surprising predictability:

Shunsaku Horiuchi (Virginia Tech) 14

Red supergiant problem

pre-image by HST SN 2008bk D ~ 4 Mpc

A deficit of high mass supernova progenitors

Kochanek et al (2008)

Log L/Lsun

Known red-supergiants (@MW, LMC) Observed progenitors

è Highest luminosity RSGs missing from progenitor sample

Smartt (2009), Smartt (2015)

Sample: volume limited to < 28 Mpc (V<2000 km/s) Pre-imaging: Limited to nearby SNe, highly successful

Shunsaku Horiuchi (Virginia Tech) 15

Red supergiant problem

Supergiants may not be exploding

• Missing RSG mass ~16 –25 Msun
• Similar to high compactness stars least likely to explode

Horiuchi et al (2014); Kochanek (2014)

Missing RSG

Smartt et al (2009)

10 20 30 40 50 60 70 80 100

initial mass [Msun] 0.1 0.2 0.3 0.4 0.5

x2.5

0.5

xcrit

(Other explanations have been explored)

• Compactness peak related to core C

burning transition.

• Exact position depends on nuclear

reaction rates, e.g., 12C(a,g)16O.

Shunsaku Horiuchi (Virginia Tech) 16

Explosions & implosions

Sukhbold & Adams (2019)

(2-parameter criterion)

Direct search candidate è Failed explosion may not be rare. Fraction can be 10–40% of all core collapse Theory:

Shunsaku Horiuchi (Virginia Tech) 17

Average neutrino flux

Average neutrino emission

• Use 100+ simulations to characterize progenitor dependence
• Use initial mass function to distribute progenitors
• Include collapse to black holes, characterized by critical compactness

5 10 15 20 25 30 35 40 45 50

E [MeV]

10

54

10

55

10

56

dN / dE [/MeV]

x2.5,crit = 0.1 x2.5,crit = 0.2 x2.5,crit = 0.3 x2.5,crit = 0.4

ne

more BH 45% BH 0% BH

Horiuchi et al (2018) e.g., papers by Fischer et al, Sumiyoshi et al, Nakazato et al, O’Connor & Ott, …

Shunsaku Horiuchi (Virginia Tech) 18

Backgrounds and search window

Optimal search window

• With water: background dominated search.
• Invisible muons: large background

Super-K (2012)

¯ νe + p → e+ + n

Signal Invisible muons

Shunsaku Horiuchi (Virginia Tech) 19

Super-K with Gadolinium

¯ νe + p → e+ + n

w/out Gd with Gd Capture on protons, signal mostly lost (~18% tagging) Capture on Gadolinium, yields a coincidence signal (~90% tagging)

Approved in 2015 Infrastructure upgrades complete Gd will be added in phases from 2019/2020 Background rejection: Signal produces a neutron, while many backgrounds do not EGADS: Evaluating Gadolinium’s Action on Detector Systems

DT ~ 30 µs, Dl ~ 50 cm

Beacom & Vagins (2004)

Shunsaku Horiuchi (Virginia Tech) 20

Backgrounds and search window

Optimal search window

• Invisible muons: largely reduced
• Gadolinium opens a signal-dominated window ~10-30 MeV

Beacom & Vagins (2004) Super-K (2012)

Shunsaku Horiuchi (Virginia Tech) 21

Upcoming sensitivities

fBH~0.2 ~0.05 ~0.03 Spectrum SK + Gd (>10MeV) [/yr] 4 MeV 1.8 +/- 0.5 4 MeV+BH(30%) ~ 2.5 SN1987A 1.7 +/- 0.5

• Sensitive to BHs
• Sensitive to core collapse history

(à tests of star formation history)

Horiuchi et al (2018); Lunardini (2009), Lien et al (2010), Yuksel & Kistler (2015), Moller et al (2018) Yuksel et al (2006)

SuperK + Gd Average emission parameters from multi-D sims similar to “D”

More BH (5 years)

Shunsaku Horiuchi (Virginia Tech) 22

MULTI-MESSENGER CONNECTIONS

Shunsaku Horiuchi (Virginia Tech) 23

+ CR

The next Galactic supernova

Nakamura, Horiuchi et al (2016)

ASAS-SN Palomar Subaru LSST LIGO VIRGO KAGRA Super-K JUNO DUNE Hyper-Kamiokande

Shunsaku Horiuchi (Virginia Tech) 24

Neutrino confidence

Reveal IF: High number statistics expected from a Galactic core collapse

Mirizzi et al (2015); Scholberg (2012)

Running Not shown: direct dark matter detectors

Shunsaku Horiuchi (Virginia Tech) 25

Alerting the community

Neutrinos: rapid sharing of core collapse occurrence Coincidence server (@BNL) Rapid ALERTs

http://snews.bnl.gov

astro-ph/0406214

ü SNEWS: ü Individual detectors

Borexino DayaBay HALO IceCube KamLAND LVD Super-K

• EGADS: automated alert within ~ 1 s
• Super-K: alert within ~ 1 hour of neutrino burst

(info: time, duration, total events, pointing)

Email, IAU, Atel, GCN

e.g., Adams et al (2013)

Shunsaku Horiuchi (Virginia Tech) 26

Multi-messenger: electromagnetic

Kistler et al (2013) based on Matzner & McKee (1999)

Shock breakout (SBO) timescales:

• For RSG (=type IIP): 1000 Rsun, 10 Msun

à days delay, hours duration

• For WR (=type Ibc): 1 Rsun, 1-10 Msun

à minutes delay, seconds duration

1hr 1day

Li et al (2010)

à Rapid alert will help chase the shock breakout signal

Shunsaku Horiuchi (Virginia Tech) 27

POINTING to core collapse Use e- scattering in the forward cone: ~300 events at SK

Pointing

Super-K Hyper-K Water only ~6 deg ~1.4 deg

Background mostly due to IBD

¯ νe + p → e+ + n ν + e− → ν + e−

Beacom & Vogel (1999), Tomas et al (2003)

Shunsaku Horiuchi (Virginia Tech) 28

POINTING to core collapse Use e- scattering in the forward cone: ~300 events at SK

Pointing

Super-K Hyper-K Water only ~6 deg ~1.4 deg Water + Gd (90% tag) ~3 deg ~0.6 deg

Background to be reduced by neutron tagging with Gd (~90% efficiency):

¯ νe + p → e+ + n

Remaining background is the ~10% of IBD and ne absorption on 16O (~20-80 events) à Pointing accuracy

• f several degrees

ν + e− → ν + e−

Beacom & Vogel (1999), Tomas et al (2003)

Shunsaku Horiuchi (Virginia Tech) 29

Multi-messenger: electromagnetic

Us

Nakamura, Horiuchi et al (2016)

Important WHERE supernova occurs: The Milky Way puts severe attenuation in the optical regime (not so much in IR)

~35% are within reach of large FOV <1m class telescopes ~20% will need need >1m class telescopes – helped by Gd pointing improvement

Shunsaku Horiuchi (Virginia Tech) 30

SNEWS2.0: Triangulation

Linzer & Scholberg (2019); See also, Brdar et al (2018), Beacom & Vogel (1999), Mhlbeier et al (2013),

Coordinated pointing: Automate triangulation into upgraded SNEWS2.0 alerts

SK+JUNO+DUNE HK+JUNO+DUNE HK+JUNO+DUNE +IceCube

Shunsaku Horiuchi (Virginia Tech) 31

Timing & Gravitational waves

WHEN one should look: e.g., IceCube: bounce time can be estimated to within ±3.5 ms at 95 % C.L.

Halzen & Raffelt (2009)

Time [ms] Frequency [Hz] 10 20 30 40 50 60 50 100 150 200 250 300 350 400 450 500 1 2 3 4 5 6 7

8.5 kpc

Time [ms] Frequency [Hz] 500 1000 200 400 600 800 1000 0.5 1 1.5 2 2.5 3 3.5 4

à Timing of core bounce helps GW detection

WITHOUT neutrino timing S/N ~ 3.5 WITH neutrino timing S/N ~ 7

Nakamura, Horiuchi et al (2016)

Shunsaku Horiuchi (Virginia Tech) 32

Theory: multiple detailed 3D hydrodynamic and oscillation simulations

Concluding remarks: future decades

Gravitational wave: Global network of GW detectors Neutrinos: rich data from present and future detectors, comparison with diffuse background Optical: high cadence telescopes

Shunsaku Horiuchi (Virginia Tech) 33

Shunsaku Horiuchi (Virginia Tech) 35

Results so far: In 7 years running,

• 6 luminous CC supernovae (09dh,

11dh, 12fh, 13ej, 03em, 14bc)

• 1 candidate failed supernova:

NGC6946-BH1 (@~6Mpc); SED well fit by 25Msun RSG

Adams et al (2016)

deficit SN2011dh V-band R-band With the candidate

Gerke et al. (2015)

è Failed fraction (90%CL)

Shunsaku Horiuchi (Virginia Tech) 37

Time-integrated neutrino signal

Horiuchi et al (2018)

Collapse to black hole speeds up with higher compactness Spectrum per core collapse Spectral parameters show systematic dependence on progenitor compactness

Shunsaku Horiuchi (Virginia Tech) 38

The challenge: beating background

H2O only H2O + Gd Search energy window [MeV] 18-30 12-38 Signal n (in 10 sec, d = 1 Mpc, 0.56Mton) ~5 ~10 Background n (over 1 day, 0.56Mton) ~0.8 ~1.1

• 1. Neutrino trigger: look for doublets or higher multiplets,

depending on bkg rate:

• Atmospheric neutrinos
• Invisible muon decays
• Spallation daughter decays

e.g., doublets in 10 sec occurs once per ~10 years (scaling SK-II to 0.56 Mton)

• 2. EM trigger: use SBO or early SN light curve to model constrain the bounce time

e.g., background rate in signal region is ~0.8 /day/0.56Mton (scaling from SK-II) à maybe even can use neutrino singlets Two approaches :

e.g., Cowen et al (2009)

Ando et al (2005)

Shunsaku Horiuchi (Virginia Tech) 40