Документ взят из кэша поисковой машины. Адрес оригинального документа : http://hea.iki.rssi.ru/conf/hea2010/Presentations/21/Levin.pdf Дата изменения: Wed Jan 26 13:25:08 2011 Дата индексирования: Tue Oct 2 00:41:17 2012 Кодировка: Поисковые слова: comet tail

. Maarten van Hoven






: - 15 Gauss 10 . Duncan & Thompson
Usov 94 Thompson et al 94-06

92

crust

· Slowly rotating, with X-ray emission powered by magnetic energy · Some magnetars also release flares

3 : 1979, 1998, 2004
Mazetz et al. 1979. Hurley et al 1998, 2005


What is the giant flare and what triggers it?
2 main possibilities: 1. Giant starquake initiated in the solid crust. vibration -> Alfven waves ->fireball
Thompson & Duncan 1995

2. Major reconnection event Slow evolution -> MHD instability -> flare
Lyutikov 03, 06

Observationally, possibility 2 is strongly preferred: few microseconds of It is possible that there are hybrids.


- (Quasi-Periodic Oscillations) Israel et al 2005


Strohmayer & Watts 06


-- !

(Israel et al 2005)


High amplitudes

courtesy Anna Watts

7


Theory of torsional vibrations of neutron stars: 1. Strong gravity: solved 2. Crustal shear modulus: uncertain by a factor of ~5

Shoemaker & Thorne 1980's Samuelsson & Karlovini 2000's Steiner & Watts 09


Theory of torsional vibrations of neutron stars:

3. Crust and core superfluids: understood a bit (a few words later) 4. Strong MHD coupling between the crust and the core: the rest of the talk


Why bother with crust-core coupling?
Fundamental core Alfven mode ~20Hz Fundamental crust shear mode ~10 ­ 40 Hz Strong coupling leads to non-trivial dynamics.

10


Torsional vibration of the whole star

L. 06, L. 07 Glampedakis et al. 06 Gruzinov 08 van Hoven & L. 2010 Gabler et al 2010 Colaiuda & Kokkotas 2010

1. Magnetically coupling to the core Pure crustal modes don't 2. Alfven continuum in the core (?). Initial What crust

on 0.01-0.1 second timescale. exist. crustal modes decay in

Crust-core dynamics:
1. Magnetically coupling to the core on 0.01-0.1 second timescale. Pure crustal modes don't exist. · Alfven continuum in the core. Initial crustal displacements decay in (L.06,07, Gruzinov 08)

· cf. Resonant absorption in solar corona (Ionson 78, Hollweg 87,
Steinolfson 85, etc.....)

Resonant Layer


Illustrative toy model

1 kg

10000 small oscillators, 0.01g


Zoom in: residual motion


Zoom in on the residual: Power spectrum: 2 Oscillations But: close to edges of the continuum


Dynamics of the continuum

amplitude

frequency


Zoom in on the edge

amplitude

frequency


Stable edge modes.
van Hoven & L 2010

There exist two discrete modes close to but outside continuum boundaries. These modes completely dominate late-time dynamics, their amplitudes are predictable from the initial data. 3 stages: 1. Exponential decay 2. Algebraic decay
(Gruzinov 08)

3. Stable edge modes


Viscous dissipation: continuum. Edge modes are long-lived.


Inflected spectrum: turning point gives strong QPO

1 kg

10000 small oscillators, 0.01g


The ``real'' axisymmetric magnetar
Basic steps: 1. Decompose crustal motion into descrete modes 2. Decompose Alfven motion on each flux surface into standing waves 3. Calculate interaction between crustal modes and Alfven standing waves. So far exact! 4. Regularize: a) Finite number of flux surfaces b) Finite number of standing waves c) Finite number of crustal modes 5. Run (e.g., leapfrog integrator) to get dynamical spectrum, or Find normal modes


The real magnetar (simulated)!


Alfven continuum. Gaps.

23


Alfven continuum. Gaps.

24


Edge modes

25


Gap modes. Location of Alfven continuum is crucial in determining the global modes.

26


Question
· So far continuum modes have lived on axisymmetric flux surfaces. How important is continuum in realistic configurations?
Answer: The more tangled is the field, the greater is the importance of discrete Alfven modes

27


Insight from simulations:

Braithwaite & Spruit 04 Braithwaite & Nordlund 06 Braithwaite 08

some conditions give nearly axisymmetric field

courtesy Jon Braithwaite
28


Insight from simulations: while others don't

Braithwaite & Spruit 04 Braithwaite & Nordlund 06 Braithwaite 08

courtesy Jon Braithwaite
29


Prescription
· Start with axisymmetric field and introduce field-line tangling which dynamically couples neighbouring flux surfaces. This is effective "magnetic"shear modulus · large tangling = large effective · The "magnetic shear-modulus coupling" collapses continuum onto dense discrete grid grid. Leads to transient QPOs!


Putting everything together: magnetar-making machine 1. 2. 3. 4. 5. Crustal modes Alfven continuum (edge and gap modes) Tangling: effective shear modulus, discrete modes' grid Mass loading in the core (superfluidity, proton fraction) Transient QPOs

Successes: 1. 20 Hz QPO as an edge mode for normal proton fraction, decoupled neutrons 2. <150 Hz QPOs are explained as edge/gap modes

31


Failure (so far): 625 Hz QPO.

32


Failure (so far): 625 Hz QPO.

33


Conclusions
· Alfven modes in the core play important role in determining the frequency and amplitude of magnetar oscillations. · This allows probing the interior of the star. Current models for QPOs at 20--100 Hz favor neutron decoupling from the oscillations (i.e. superfluidity) · Current (incomplete survey of) models struggle with 625Hz QPO. This may imply even more radical decoupling (or oscillations of the magnetosphere itself). Possibilities include
a. much smaller proton fraction than currently thought b. magnetic field largely confined to the crust c. Unusual QCD phase: color-flavour-locked superconductor
34