Resonant Neutrino Self-Interaction

with U(1)Lμ-Lτ Model

AYUKI KAMADA (UC Riverside) in collaboration with Hai-bo Yu (UCR)

@第28回宇宙ニュートリノ研究会

“Dip” in IceCube

Good fit w/ ϕ(Ε)∝E-2

Missing events

at subPeV M. Aartsen et al. 2014

Possible InterpretationSources w/ ϕ(Ε)∝E-2

• Astrophysical (GRB, SNR)• Cosmogenic (pp, pγ…)• New Physics (DM decay, annihilation)

Origins w/ dip• Statistical fluctuation

-a few events expected (2σ consistent)• New Physics (neutrino interactions)

This talk

How does neutrino self-interaction work?

𝛎p~PeV

𝛎

k~10-4 eV

𝛎

𝛎

p’~p/2

k’~p/2

High energy neutrino

cosmic neutrinobackground (C𝜈Β)

High energy neutrino deposits their energy in cosmic neutrino background

Exclusive for subPeVOnly subPeV neutrinos interact effectively

achieved by Breit-Wigner Resonance

cosθ: injection angleTypically m𝜈 ~ 0.1 eV

>> k~10-4 eV~10 MeV

𝜈𝜈 -> Z’ -> 𝜈𝜈

mZ0 =p2m⌫p

m2Z0 = 2p(

pk2 +m2

⌫ � k cos ✓)

10 MeV mediator implied!!

Well-motivated model

Gauging (muon number - tau number) 𝜈μ(+1), 𝜈τ(-1)… anti-particles(opposite sign)

couple with Z’-boson (mZ’~10MeV!!)

Advantages• Anomaly-free (well-defined theory)• natural explanation of maximal mixing

for atmospheric neutrino θ23~π/4• No experimental difficulty

unlike U(1)Le-Lμ, U(1)Le-Lτ

(e.g. beam-dump experiment)

U(1)Lμ-Lτ

Muon g-2

aμ=(g-2)/2~0.001

Measured muon anomalous magnetic moment (g-2) deviates from standard model prediction

𝛥aμ(exp)=aμ(exp) - aμ(SM)=(42.6±16.5)×10-10

Z’-boson exchange contribution

�a

Z0

µ =g

0 2

8⇡2

Z 1

0dx

2m2µx

2(1� x)

x

2m

2µ + (1� x)m2

Z0

Gauge coupling of g’~10-4 can fill the gap!!

Optical depth

e⌧s(z) = 18⇣(3)T 3⌫,0(1 + z)3

1

H(z)m2Z0

�Z0

mZ0

' 1

✓g0

2.5⇥ 10�5

◆2

C𝜈Β number density

Streaming Length

Cross section

Gauge coupling of g’~10-4 is also sufficient to explain dip in IceCube!!

T. Araki et al. 2014

𝛎Β

𝛎Β

Is this brilliant story true?𝛎e

𝛎Β

𝛎μ

𝛎τ

𝛎e 𝛎Β

𝛎

𝛎

𝛎

𝛎

𝜈e’s can reach us without losing energy flavored (<->flavor-blind)

Neutrino oscillation I𝛎e

Starting with

IceCube

w/o oscillation𝛎Β

𝛎Β 𝛎e

Source 𝛎e

𝛎e 𝛎μ

𝛎τ

Neutrinos change their flavors during

oscillations

(c) IceCube

(c) NASA

Neutrino oscillation II𝛎e

Starting with

IceCube

w/ oscillation𝛎Β

𝛎Β 𝛎e

Source 𝛎e

𝛎μ

𝛎τ

𝛎Β

𝛎Β

Even starting with 𝜈e’s, only a part of 𝜈’s can reach us

(c) NASA

(c) IceCube

Neutrino oscillation IV

P '

2

40.30 0.13 0.120.13 0.06 0.050.12 0.05 0.04

3

5

+e�e⌧s(zs)/2

2

40.07 �0.05 �0.03�0.05 0.03 0.02�0.03 0.02 0.01

3

5

+e�e⌧s(zs)

2

40.18 0.15 0.120.15 0.29 0.310.12 0.31 0.35

3

5

Starting with

Probability matrix𝛎e 𝛎μ 𝛎τ End up with

𝛎e

𝛎μ

𝛎τ

Predictions and Prospectspp interaction(�⌫e ,�⌫µ ,�⌫⌧ ) ' (1, 2, 0)

@ IceCube (1/3 survived)(0.56, 0.25, 0.22)

z=zs z=0 Accumulated date can

distinguishflavored modelfrom

flavor-blindmodel

Cosmological/Astrophysical Aspects

Cosmology of 10 MeV Z’-boson I

inverse decay pair creation

decay pair annihilationTwo relevant processes

Dominant @ high-T @ low-TTcrit ' 100GeV ⇥

✓10�4

g0

◆⇥

⇣ mZ0

10MeV

⌘

Cosmic history w/ Z’

T~10 MeV T~0.1 MeVT~1 MeV

e 𝜈 𝛾

𝜈

e 𝛾 𝛾

BBNe-e+ annihilation𝜈’s decoupling

Ζ᾽

Ζ᾽

μ/π’s decay

Important question: When is Z’ decays?

Decoupling of Z’ III

T~10 MeV T~0.1 MeVT~1 MeV

e 𝜈 𝛾

𝜈

e 𝛾

BBNe-e+ annihilation𝜈’s decoupling

Ζ᾽

Ζ᾽

μ/π’s decay

Z’ decays in-between

heated 𝜈’s change the expansion ratespoil success of SM in BBN

𝛾

mZ’> 6.3 MeV

Core Collapse SNCore

Produced 𝜈’s can not escape from proto-neutron star

Neutrino sphere

Random walk

R = λ√N

T~8 MeVR~15 km

mean free path # of collisions

N=t/λλ~4 mt~10s

It takes

until escape

λ

𝛎

SN1987AThis 10s determines the duration of neutrino cooling

M. Nakahata 2007

Core Collapse SN w/ Z’Core

Same story for 𝝂e, but not for 𝝂μ 𝝂τ

Neutrino sphere

𝝂e

T~8 MeVR~15 km

inverse decay

decay𝛎e𝛎μ

Diffusion time w/ Z’

Diffusion time t~105s >> 10s (SN1987A)Mean free path λ~0.01cm

For g’~10-4 (good for Muon g-2, IceCube dip)

Notion• More detailed study needed to conclude…• Some additional cooling mechanism of core

(e.g. axion, hidden photons)

Only 𝝂e’s contribute to SN cooling

Prediction

𝛎e𝛎μ

SN

Super Kamiokande

𝛎τOnly 𝝂e’s are emitted from SN

Important hint ofthis scenario

(c) NASA

Summary