Why supernova neutrino oscillations are fun and why three-flavor - - PowerPoint PPT Presentation

▶
why supernova neutrino oscillations are fun
SMART_READER_LITE
LIVE PREVIEW

Why supernova neutrino oscillations are fun and why three-flavor - - PowerPoint PPT Presentation

Nov 02, 2022 120 likes •572 views

Why supernova neutrino oscillations are fun and why three-flavor analysis is a must Alex Friedland, LANL INT neutrino workshop, Feb 10, 2010 Based on A.F ., 1001.0996 + in prep. Movies, etc, are at

slide-1
SLIDE 1

Why supernova neutrino

  • scillations are fun

and why three-flavor analysis is a must Alex Friedland, LANL

INT neutrino workshop, Feb 10, 2010 Based on A.F ., 1001.0996 + in prep.

Movies, etc, are at http://alexfriedland.com/papers/supernova/latecollective/

1 Wednesday, February 10, 2010

slide-2
SLIDE 2

Supernova neutrinos: key to a big puzzle

Supernova explosions are some of the most important processes in the Universe that influenced our world

“Every one of our chemical elements was once inside a star. The same

  • star. You and I are brothers. We came from the same supernova.”

From the NYTimes obituary for Geoffrey Burbidge, Feb 6, 2010

Simulations of the galactic disk seem to show the supernova feedback crucial to its structure. Neutrinos come to us straight from the central engine, r ~ 10 01 km. Could provide the resolution of the 50-year old puzzle -- how the massive stars

  • explode. Unlike SN1987a, 10

04 events -- second-by-second spectra

2 Wednesday, February 10, 2010

slide-3
SLIDE 3

Flavor transformations

By now, we know that neutrinos oscillate between flavors solar, atmospheric, reactor, beam Supernova neutrinos must also transform flavors, no longer a choice To extract physics from the signal, these transformations must be understood!

3 Wednesday, February 10, 2010

slide-4
SLIDE 4

Neutrino oscillations: simple always works?

4 Wednesday, February 10, 2010

slide-5
SLIDE 5

Neutrino oscillations: simple always works?

Neutrinos are already very hard to detect, do Nature decided to be kind to us.

4 Wednesday, February 10, 2010

slide-6
SLIDE 6

Neutrino oscillations: simple always works?

Neutrinos are already very hard to detect, do Nature decided to be kind to us. In all the known cases, we got lucky: things are simpler than they could’ve been atmospheric neutrinos s: 2-flavor oscillations, 2E/ /m2 ~ 10 04 km for E ~ 1 GeV and m2 ~ 3 * 10 0-3 eV V2 solar neutrinos s: again 2-flavor oscillations, m2/2E ~ G GF n ne for E ~ 1 MeV and ne in the solar center KamLAND D: again 2 flavors, 2E/ /m2 ~ 10 02 km for E ~ 10 MeV and m2 ~ 8 * 10 0-5 eV MINOS S: again 2 flavors. In fact, we are trying hard to see the 3-flavor effects

4 Wednesday, February 10, 2010

slide-7
SLIDE 7

Neutrino oscillations: simple always works?

Neutrinos are already very hard to detect, do Nature decided to be kind to us. In all the known cases, we got lucky: things are simpler than they could’ve been atmospheric neutrinos s: 2-flavor oscillations, 2E/ /m2 ~ 10 04 km for E ~ 1 GeV and m2 ~ 3 * 10 0-3 eV V2 solar neutrinos s: again 2-flavor oscillations, m2/2E ~ G GF n ne for E ~ 1 MeV and ne in the solar center KamLAND D: again 2 flavors, 2E/ /m2 ~ 10 02 km for E ~ 10 MeV and m2 ~ 8 * 10 0-5 eV MINOS S: again 2 flavors. In fact, we are trying hard to see the 3-flavor effects General principle e: experimental results must conveniently fit in the PRL format t?

4 Wednesday, February 10, 2010

slide-8
SLIDE 8

SN neutrinos: complexity returns with a vengeance!

Two resonant densities (solar and atm.) Neutrinos and antineutrinos of all flavors Density profile changing with time: Shock, turbulence Neutrino-neutrino interactions (coherent forward scattering) ...

5 Wednesday, February 10, 2010

slide-9
SLIDE 9

Progress?

Ten years ago, we had a definite prediction for what the supernova signal looks like The field has changed radically, as it was shown that many new effects are important for SN neutrino

  • scillations

We do not have a clear prediction anymore Will complexity render the supernova signal useless?

6 Wednesday, February 10, 2010

slide-10
SLIDE 10

Yes, progress! Why “complexity” is good

The “complexity” actually makes the signal more useful, not less useful, as it provides new ways the information about the developing explosion can be imprinted in the neutrino signal

7 Wednesday, February 10, 2010

slide-11
SLIDE 11

Concrete example

Shock and turbulence

  • R. Schirato & G. Fuller (2002)
  • A. F. & A. Gruzinov (2006)

8 Wednesday, February 10, 2010

slide-12
SLIDE 12

3D simulations

3d simulations of the accretion shock instability Blondin, Mezzacappa, & DeMarino (2002) See e http://www.phy.ornl.gov/ tsi/pages/simulations.html No central heating. Still, extensive, well-developed turbulence behind the shock

9 Wednesday, February 10, 2010

slide-13
SLIDE 13

3D simulations

3d simulations of the accretion shock instability Blondin, Mezzacappa, & DeMarino (2002) See e http://www.phy.ornl.gov/ tsi/pages/simulations.html No central heating. Still, extensive, well-developed turbulence behind the shock

9 Wednesday, February 10, 2010

slide-14
SLIDE 14

3D simulations

A beautiful simulation from the web page of K.Kifonidis http://www.mpa- garching.mpg.de/~kok/ Neutrino flavor transformations happen in the dynamically changing profile of the expanding shock and turbulence

10 Wednesday, February 10, 2010

slide-15
SLIDE 15

3D simulations

A beautiful simulation from the web page of K.Kifonidis http://www.mpa- garching.mpg.de/~kok/ Neutrino flavor transformations happen in the dynamically changing profile of the expanding shock and turbulence

10 Wednesday, February 10, 2010

slide-16
SLIDE 16

Signatures of shock and turbulence

Time-varying features sweep through the spectrum several seconds after the onset of explosion The development of the explosion may be observed in real time!

11 Wednesday, February 10, 2010

slide-17
SLIDE 17

Signatures of shock and turbulence

Time-varying features sweep through the spectrum several seconds after the onset of explosion The development of the explosion may be observed in real time!

11 Wednesday, February 10, 2010

slide-18
SLIDE 18

Core-collapse supernova and convection

Convection behind the shock front is not just a curiosity: essential for the explosion mechanism! ( (Herant, Benz, Hix, Fryer, Colgate Ap. J. 435, 339 (1994) ))

Scheck, Plewa, Janka, Kifonidis, and Muller,

  • Phys. Rev. Lett. 92, 011103 (2004), t=1 s

Convection brings energy from the dense region near the proto-neutron star to the region behind the shock Observing it would confirm the basic ingredient in the current paradigm of the SN explosion

12 Wednesday, February 10, 2010

slide-19
SLIDE 19

Collective oscillations

Close to the protoneutron star, the neutrino background itself becomes important in the oscillation Hamiltonian The neutrino induced contribution is proportional to the neutrino density matrix, has off-diagonal components The problem becomes s non-linear r: changing the neutrino states also changes the background that drives the evolution Recently, detailed numerical calculations of this effect were performed by Duan, Fuller, Carlson, Qian, 2005, 2006 (and followed by others) that led to a realization that complex flavor transformations occur at ~ 100-300 km

√ 2GF

  • i

ni(1 − cos θij)|ψiψi|

Pantaleone, 1992

13 Wednesday, February 10, 2010

slide-20
SLIDE 20

Collective oscillations

14 Wednesday, February 10, 2010

slide-21
SLIDE 21

Collective oscillations

Collective effects operate simultaneously with shock/ turbulence Just like turbulence/shock, collective effects rely only

  • n known physics

Need to be computed, no way out!

14 Wednesday, February 10, 2010

slide-22
SLIDE 22

Collective oscillations

Collective effects operate simultaneously with shock/ turbulence Just like turbulence/shock, collective effects rely only

  • n known physics

Need to be computed, no way out! There has been an avalanche of papers on this subject in the last several years

14 Wednesday, February 10, 2010

slide-23
SLIDE 23

Collective oscillations

Collective effects operate simultaneously with shock/ turbulence Just like turbulence/shock, collective effects rely only

  • n known physics

Need to be computed, no way out! There has been an avalanche of papers on this subject in the last several years Plan A: extract results from existing literature

14 Wednesday, February 10, 2010

slide-24
SLIDE 24

Collective oscillations

Collective effects operate simultaneously with shock/ turbulence Just like turbulence/shock, collective effects rely only

  • n known physics

Need to be computed, no way out! There has been an avalanche of papers on this subject in the last several years Plan A: extract results from existing literature

14 Wednesday, February 10, 2010

slide-25
SLIDE 25

Existing literature

Existing 3-flavor calculations are done with different fluxes and spectra

  • a. Since the

problem is nonlinear, different initial conditions may give different results. Indeed, calculations with the late-time spectra of the type we are interested in seem to give very curious, novel results, including multiple spectral swaps This can be potentially very significant But these calculations are done with only 2 flavors. Is the third state a spectator?

Antineutrinos IH Neutrinos IH 10 20 30 40 Energy [MeV] NH 10 20 30 40 50 Energy [MeV] NH

Dasgupta, Dighe, Raffelt, Smirnov, arXiv:0904.3542 [hep-ph] -> PRL (2009)

15 Wednesday, February 10, 2010

slide-26
SLIDE 26

Start by repeating the 2- flavor calculations

Complete agreement with 0904.3542

This, and subsequent movies are at http://alexfriedland.com/papers/supernova/latecollective/

16 Wednesday, February 10, 2010

slide-27
SLIDE 27

Repeat but with all 3 flavors

17 Wednesday, February 10, 2010

slide-28
SLIDE 28

Repeat but with all 3 flavors

17 Wednesday, February 10, 2010

slide-29
SLIDE 29

Repeat but with all 3 flavors

Entirely different result!?

17 Wednesday, February 10, 2010

slide-30
SLIDE 30

How can this be?

The high-energy spectral split is gone How can the solar splitting, which is only 1/30 of the atmospheric

  • ne, undo the effects of

the latter? For the antineutrinos s, the result is also interesting. The spectral swap is only partial ( (mixed spectrum m)

10 20 30 40 50 60 0.5 1. 1.5 2. 2.5 3. neutrino energy MeV

Inverted Hierarchy, 2 flavors, Νe

initial Νx initial Νe Νe at 1000 km 10 20 30 40 50 60 0.5 1. 1.5 neutrino energy MeV

Inverted Hierarchy, 2 flavors, Νe

initial Νx initial Νe Νe at 1000 km 10 20 30 40 50 60 0.5 1. 1.5 2. 2.5 3. neutrino energy MeV

Inverted Hierarchy, full 3 flavors, Νe

initial Νx initial Νe Νe at 1000 km 10 20 30 40 50 60 0.5 1. 1.5 neutrino energy MeV

Inverted Hierarchy, full 3 flavors, Νe

initial Νx initial Νe Νe at 1000 km

18 Wednesday, February 10, 2010

slide-31
SLIDE 31

Let’s at what m2 the differences kick in

At first glance, this result is strangest of all: At m2=0, 2-flavor result is reproduced As soon as m20, the answer is closer to the realistic m2 than to m2=0

10 20 30 40 50 60 0.5 1. 1.5 2. 2.5 3. neutrino energy MeV

Νe at 1000 km, different m

2 initial Νx initial Νe m

2 0.5std. val.

m

2 0.2std. val.

m

2 0.01std. val.

m

2 0 19 Wednesday, February 10, 2010

slide-32
SLIDE 32

The answer: instabilities, instabilities

Recall that the concept of instability is key to understanding collective

  • scillations

The initial configuration is nearly in flavor eigenstates, yet the large flavor mixing develops later The key role of this instability for supernova neutrinos was only understood in Duan, Fuller, Qian, astro-ph/0511275 (and the fact that dense matter doesn’t suppress it) Interestingly, the simple system of two angular momenta is not merely analogous to an inverted pendulum, but in fact turns out to be exactly like it (Hannestad, Raffelt, Sigl, Wong, astro-ph/0608695)

“However, the nonlinearity of the system creates an instability ... The situation is analogous to a rigid pendulum positioned [...] suddenly inverted.” Kostelecky & Samuel, Neutrino oscillations in the early universe with an inverted neutrino mass hierarchy y, PLB 318, 127 (1993).

20 Wednesday, February 10, 2010

slide-33
SLIDE 33

Cartoon analogy

The 2-flavor instability can be (schematically) pictured like this

21 Wednesday, February 10, 2010

slide-34
SLIDE 34

Cartoon analogy

The 2-flavor trajectory is itself unstable in the 3-flavor space Disclaimer: only a cartoon, so that I don’t have to draw 8-dim.

  • bjects! (SU(3))

22 Wednesday, February 10, 2010

slide-35
SLIDE 35

Mass basis

All three states are seen to participate in the “bipolar- like” oscillations, hence trajectory is very different from 2 flavors Atmospheric decouples at ~ 200 km, solar at ~500 km Eventually, a single split is formed between 2 and 3.

400 600 800 1000 radius km eigenvalues

Ν2 Ν1 Ν3 Inverted, neutrinos

200 400 600 800 1000 0.5 1. radius km

PΝeΝ1

200 400 600 800 1000 0.5 1. radius km

PΝeΝ2

200 400 600 800 1000 0.5 1. radius km

PΝeΝ3

23 Wednesday, February 10, 2010

slide-36
SLIDE 36

Instability, zoomed in

The instability first grows between 2 and 3, but then also between 3 and 1. The growth rates are the same, indicating that the atmospheric splitting is the driving force in both. The solar splitting provides the initial mixing, just to kick off the instability Analogous to the role of small 13 for seeding the 2-flavor instability

40 50 60 70 80 90 100 110 1012 1010 108 106 104 0.01 1 radius km

PΝeΝ3

40 50 60 70 80 90 100 110 1014 1011 108 105 0.01 radius km

PΝeΝ1

400 600 800 1000 radius km eigenvalues

Ν2 Ν1 Ν3 Inverted, neutrinos

24 Wednesday, February 10, 2010

slide-37
SLIDE 37

Antineutrinos: mixed spectrum

The would-be split that is driven by the solar splitting is never completed in this case: adiabaticity violated The scale height of the nu-nu potential is The atmospheric distance scale is safely adiabatic while the solar scale is only marginally adiabatic

400 600 800 1000 radius km eigenvalues

Ν3 Ν1 Ν2 Inverted, antineutrinos

200 400 600 800 1000 0.5 1. radius km

PΝeΝ1

200 400 600 800 1000 0.5 1. radius km

PΝeΝ2

200 400 600 800 1000 0.5 1. radius km

PΝeΝ3

p , |d ln Hνν/dr|−1 ∼ r/4 ∼ 75 − 100 km

2E/∆m2

atm ∼ 2 km,

2E/∆m2

⊙ ∼ 77 km,

25 Wednesday, February 10, 2010

slide-38
SLIDE 38

More on adiabaticity

To illustrate this argument, let’s artificially increase the solar splitting broad split forms The adiabaticity is reflected in the width of a

  • split. (See Raffelt &

Smirnov, 2007) Our finding can be viewed as an extremely broad split.

10 20 30 40 50 60 0.5 1. 1.5 neutrino energy MeV

Νe at 500 km, different m

2 initial Νx initial Νe m

2 10std. val.

m

2 5std. val.

m

2 std. val. 26 Wednesday, February 10, 2010

slide-39
SLIDE 39

Normal hierarchy

The high-energy splits persist in this case Solar non-adiabaticity is present

27 Wednesday, February 10, 2010

slide-40
SLIDE 40

Normal hierarchy

The high-energy splits persist in this case Solar non-adiabaticity is present

27 Wednesday, February 10, 2010

slide-41
SLIDE 41

Robustness: decrease luminosity

Results don’t change qualitatively -- reassuring

28 Wednesday, February 10, 2010

slide-42
SLIDE 42

Robustness: decrease luminosity

Results don’t change qualitatively -- reassuring

28 Wednesday, February 10, 2010

slide-43
SLIDE 43

Conclusions

Unlike 10 years ago, we no longer have a simple prediction for the supernova neutrino signal... ... Because the physics turned out to be much more interesting! The ingredients are all known physics, so must sort out what’s going on Genuine 3-flavor effects; non-factorizable Potentially unique information about the physics of the explosion + probes of neutrino properties probes of new physics -- not yet understood

29 Wednesday, February 10, 2010

slide-44
SLIDE 44

P .S. nonlinearity at work, expect more surprises?

Even simple systems of nonlinear ODEs can have very rich and nontrivial behavior

dx dt = σ(y − x) dy dt = x(ρ − z) − y dz dt = xy − βz

30 Wednesday, February 10, 2010