High-Energy Neutrinos from Supernovae: New Prospects for the Next Galactic Supernova

Transcription

1 High-Energy Neutrinos from Supernovae: New Prospects for the Next Galactic Supernova arxiv: Kohta Murase (Penn State) TeVPA Columbus, Ohio

2 Neutrinos: Unique Probe of Cosmic Explosions Super-K ~10 MeV neutrinos from supernova thermal: core s grav. binding energy - supernova explosion mechanism - progenitor - neutrino properties, new physics Super-K can detect ~8,000 n at MeV (at 8.5 kpc) IceCube GeV-PeV neutrinos from supernova? non-thermal: shock dissipation - physics of cosmic-ray acceleration - progenitor & mass-loss mechanism - neutrino properties, new physics IceCube/KM3Net can detect??? n at TeV

3 Neutrinos: Unique Probe of Cosmic Explosions Super-K ~10 MeV neutrinos from supernova thermal: core s grav. binding energy - supernova explosion mechanism - progenitor - neutrino properties, new physics Super-K can detect ~8,000 n at MeV (at 8.5 kpc) IceCube GeV-PeV neutrinos from supernova? non-thermal: shock dissipation - physics of cosmic-ray acceleration - progenitor & mass-loss mechanism - neutrino properties, new physics IceCube/KM3Net can detect ~1000 n at TeV

4 Early Diffusive Shock Acceleration in Supernovae? supernova remnant (Cas A) CR and high-energy neutrino production is initially negligible most of energy is in a kinetic form until the Sedov time uniform ISM: CR energy dissipation energy t 3 But situations are different when circumstellar material (CSM) exists many observational evidences in the recent several years (Raffaella Margutti s talk)

5 Evidence of Strong Interactions w. Dense CSM SN 2010jl (IIn) Fransson et al. SN 2014C (Ib->IIn) Margutti et al. 16 L t 3/(n 2), n = 7.6 log L bol (erg s 1 ) nent s in ure ly is ly ust the Figure 6. Bolometric light curve for the SN ejecta, excluding the dust echo. The dashed lines show power-law fits to the early and late light curves used to construct the density distribution of the explosion. Note the pronounced break Ofek+14 ApJ in the light curve at 320 days. The dashed lines give power-law fits to the luminosity before and after the break (see Section 4.5 for a discussion). (A color version of this figure is available in the online 91 journal.) Flux [erg cm 2 s 1 Å 1 ] Fransson+14 ApJ log(days after first detection) (t/100 days) erg s 1 and a final steep decay L(t) = (t/320 days) 3.39 erg s 1 after day Ofek et al. (2014) estimatethebolometriclightcurveby assuming a constant bolometric correction of 0.27 mag to the R-band photometry. With this assumption, they find a flatter light curve with L(t) t 0.36 for the same explosion date as we use here. The reason for this difference is that the R-band decays slower than most of the other bands, 978 as can be seen from Figure 2. Thebolometriclightcurvewillthereforebesteeper than the R-band light curve. OIII OII Hγ Hβ HeI OI Hα HeI CaII The slope depends on the6000 assumed 7000shock 8000 breakout 9000 date Ofek et al. (2014)discussthisbasedonthelightcurveandfindalikely Rest wavelength [Å] range of days before I-band maximum, corresponding to JD 2,455,4692,455,479. Using 2,455,469 instead of our 120 examples of strong interactions w. dense wind or CSM (IIn, SLSN-II) FIG. 10. This plot summarizes the key and unique observational features of SN 2014C over the electromagnetic spectrum. Central Panel: X-ray (red and radio (7.1 GHz, blue stars) evolution of SN 2014C compared to a sample of Ibc SNe from Margutti et al. (2014b) and Soderberg et al. (2010). SN shows an uncommon, steady increase in X-ray and radio luminosities until late times, a signature of the continued shock interaction with very dense m in the environment. Upper panels: The optical bolometric luminosity of SN 2014C is well explained at early times by a model where the source of ene

6 Evidence for Dense Material in Ordinary SNe II early spectroscopy (Yaron+ 16 Nat. Phys.) SN 2013fs see also light curve modeling Morozova+ 17 ApJ Extended material is common even for Type II-P SNe Mdot~ M sun yr -1 (>> 3x10-6 M sun yr -1 for RSG)

7 Supernovae with Interactions with CSM wind/shell wind/shell ejecta Star SN shocks kinetic energy thermal + non-thermal via shock p + p Nπ + X π ± ν µ +ν µ + ν e (ν e ) + e ± π 0 γ +γ dense environments = efficient n emitters (calorimeters)

8 Shock Dynamics -> Time-Dependent Model equation of motion of the shocked ejecta M sh dv s dt =4 R2 s[% ej (V ej V s ) 2 % cs (V s V w ) 2 ] self-similar solution (Chevalier 82) shock radius R s = X(w, )D 1 w E 3 2( w) ej M ej 5 2( w) t 3 w CSM parameter D = Ṁw 4 V w w=2 for a wind CSM E ej ~ erg, M ej ~ 10 M sun d~10-12 for typical progenitors dissipation luminosity L d =2 % cs Vs 3 Rs 2 6w 15+2 w / t w parameters can be determined by photon (opt, X, radio) observations! E d ~ E ej (>V s ) in the detailed model, different from E d ~(M cs /M ej )E ej by KM+11

9 Diversity of Core-Collapse Supernovae Wheeler 90 II-P 48.2% II-L 6.4% Ic 14.9% IIn 8.8% Smith+ 11 MNRAS Ib 7.1% Ibc-pec 4.0% IIb 10.6% Core-Collapse SN Fractions Type II SN frac. ~ 2/3 Class D Ṁ w [M yr 1 ] V w [km s 1 ] R [cm] IIn II-P a II-P b II-L/IIb Ibc w. pre-sn mass loss stellar wind only

10 log(e L E [erg s -1 ]) Neutrino Light Curve IIn II-P II-L/IIb Ibc w. pre-sn mass loss w.o. pre-sn mass loss t onset t [s] t onset ~ time leaving the star (typical) or breakout time (IIn) slowly declining light curve while pion production efficiency ~ 1

11 log(e 2 [GeV cm -2 ]) all flavor d=10 kpc Neutrino Fluence IIn II-P II-L/IIb Ibc E [GeV] thick: s=2.2 thin: s=2.0 Fluence for an integration time at which S/B 1/2 is maximal (determined by the detailed time-dependent model)

12 Prospects for Neutrino Detection ) log(n µ <t IIn II-P II-L/IIb Ibc background (atm. + astro) t [s] through-going muon events d=10 kpc ~ events for Type II supernovae at 10 kpc ~ events for Ibc (but see Kashiyama, KM+ 13 ApJL)

13 Some Remarks Testable & clear predictions (no need for jets, winds, shocks in a star) free parameters: e CR & s shock acceleration theory (e CR ~0.1 & s~ ) Time window depends on SN types; guidelines are provided by the theory (f pp ~t dyn /t pp ~1) e.g., characteristic time window: ~1-10 day for SNe II Energy range IceCube/KM3Net: TeV-PeV (detectable Glashow res. anti-n e & n t events) Hyper-K/PINGU/ORCA: GeV * Type II cases are different from the Type IIn case II-P/II-L/IIb/Ibc: shock in the CSM is collisionless & M csm << M ej IIn: shock can be radiation-mediated & M csm could be larger than M ej t onset determined by photon breakout, ejecta deceleration, radiative shock, other relevant CR cooling processes (for work on SNe IIn, see KM, Thompson, Lacki & Beacom 11 and Petropoulou s talk)

14 Implications Astrophysical implications a. pre-explosion mass-loss mechanisms how does a dense wind/shell form around the star? b. PeVatrons can CRs be accelerated up to the knee energy at ev? c. real-time observation of ion acceleration for the first time is it consistent with the diffusive shock acceleration? d. promising targets of multi-messenger astrophysics MeV ns & possibly gravitational waves optical, X-rays, radio waves, and gamma rays (up to ~Mpc by Fermi) Particle physics implications neutrino flavors (matter effect is not relevant), neutrino decay, neutrino self-interactions, oscillation into other sterile states etc. cf. more lucky examples? Betelgeuse: ~10 3-3x10 6 events Eta Carinae: ~10 5-3x10 6 events

15 Take Away - We provided the time-dependent model for high-energy neutrino/gamma-ray emission from different classes of SNe - Type II: ~1000 events of TeV n from the next Galactic SNe - SNe as multi-messenger & multi-energy neutrino source L n MeV n GeV-PeV n ~10 sec ~0.1-1 day E n

16 CRs Should Lead to Efficient Hadronic Interactions particle collisions with CSM t pp = 1/(n k pp s pp c) f t dyn = R/bc pp = (R/b) n k pp s pp f pp (r bo ) ~ b -2 (k pp s pp /s T ) ~ 0.03 b -2 p + p Nπ + X (s pp ~3x10-26 cm 2 ) at breakout: t T =1/b b ~ transrelativistic SNe b ~ nonrelativistic SNe most CR energy goes to pions π 0 γ +γ π ± ν µ +ν µ + ν e (ν e ) + e ± new probes!

17 Transrelativistic SNe (b~1) dh?? 100 ms pulse Cosmological GRB?? $= lw? XRF020903? 2006aj? R= bw? $#~1 2008D! #=0.1 Fan et al. 11 ApJL from Katz 11 Nearby GRBs (ex. may form another class much dimmer (E iso LLg ~10 50 erg E iso GRBg ~10 53 erg/s ) more frequent (r LL ~ Gpc -3 yr -1 r GRB ~ Gpc -3 yr -1 ) relativistic ejecta (the other GRB-SNe bb) (Soderberg+ 10 Nature) - maybe more baryon-rich? (e.g., Zhang & Yan 11 ApJ)

18 Neutrinos from Transrelativistic SNe log(e ν 2 φν [erg cm -2 ]) sec 1day Kashiyama, KM+ 13 ApJL Total pγ pp ANB log(e ν [GeV]) Detectable by IceCube up to ~10 Mpc Stacking analyses may be possible

19 Gamma Rays from Transrelativistic SNe log(e γ 2 Φγ [erg cm -2 sec -1 ]) Mpc log(e γ [GeV]) 100Mpc Detectable by CTA up to ~100 Mpc attenuated injected CTA (0.5hr) Kashiyama, KM+ 13 ApJL Detection SN-GRB connection, physics of breakout and CRs

20 Shock Breakout Emission from Dense CSM photon diffusion time t diff ~ R 2 /k rad ~ns T R 2 /c dynamical time: t dyn ~ R/bc, b=v/c (a) shock breakout: t rise =t diff =t dyn t T =1/b=c/V CSM mass: M cs ~ (4pR 2 /3s T )m P t T Dissipation energy: E rad ~(1/2)M cs V 2 Chevalier & Irwin 11 ApJL ex. SN 2009ip t rise =10 d, R=0.5x10 15 cm M cs ~0.05 M sun E rad ~ 2x10 49 erg consistent w. obs.!

21 Neutrinos from Type IIn SNe Dt= s KM, Thompson, Lacki & Beacom 11 PRD Dt=10 7 s Model A - optically thick collision Model B - optically thin collision Atm. n µ within 1 d=10mpc If CRs carry ~10% of E ej # of µs ~a few for SN@10Mpc Stacking analyses for nearby SNe (~O(100) needed)

22 Gamma Rays from Type IIn SNe KM, Thompson, Lacki & Beacom 11 PRD log(e γ 2 Fγ [GeV cm -2 s -1 ]) Fermi (t= s) Fermi (t= s) Model A Model B (RS) Model B (FS) CTA (100 hr) Model A - optically thick collision Model B - optically thin collision -13 d=10mpc log(e γ [GeV]) GeV g rays can be seen by Fermi up to ~30 Mpc TeV g rays are detectable by CTA up to ~ Mpc