1
Nanda Rea Institut de Ciencies de l’Espai,
CSIC-IEEC, Barcelona
The neutron star zoo
A bit of history
• 1931 Chandrasekhar argued that WDs collapse at masses > 1.4 M�. (Chandrasekhar 1931, ApJ)
• 1934 Baade & Zwicky proposed the existence of NS, they predicted their formation due to supernova explosion and their radius of ~10 km . (Baade & Zwicky 1934, Proc.Nat.Acad.Sci.)
• 1939 Oppenheimer & Volkoff delined the first equation of state for a NS of mass ~1.4 M�, a radius of ~10 km and a density of ~1014 gr/cm3 (Oppenheimer & Volkoff, Phys.Rev)
• 1967 Pacini predicted electromagnetic waves from rotating NSs and that such star might be powering the Crab nebula. (Pacini 1967 and 1968, Nature)
• 1968 Hewish & Bell studing interplanetary scintillation observed a periodicity of 1.337s, discovering the first pulsar: PSR 1919+21. (Hewish et al. 1968, Nature)
Nanda Rea CSIC-IEEC
!
˙ E rot = "2
3c 3 ˙ ̇ m 2 = "2B2R6#4 sin2$
3c 3
!
P ˙ P = 8" 2Rns6
3c 3I
#
$ %
&
' ( B0
2 sin2)
Radio pulsars: characteristics
Nanda Rea CSIC-IEEC
4
Magnetic Field Tree
0.6 G – The Earth magnetic field measured at the North pole
100 G – A common hand-held magnet like those used to stick papers on a refrigerator
107 G – The strongest man-made field ever achieved, made using focussed explosive charges, lasting only 4-8 s
1012 G – Typical neutron star magnetic fields
4.4x1013 G – Electron critical magnetic field
10 14 -1015 G: Magnetars overtake this limit…
Unique places to study the physics of plasma embedded in very high magnetic and gravitational fields
Nanda Rea CSIC-IEEC
1- Radio pulsars: ~2000
4- X-ray Isolated Neutron Stars (XINSs): ~9
5- Rotating Radio Transients (RRATs): ~70
3- X-ray binary neutron stars: ~1000
7- Magnetars (AXPs & SGRs): ~20
Different classes of neutron stars
2- Binary pulsars: ~60
Nanda Rea CSIC-IEEC
6- Central Compact Objects (CCOs): ~5
Pulsar Wind Nebulae
Radio pulsars: multiband emission
Radio Optical Soft X Hard X >200 MeV >5 GeV
- Particle acceleration - synchrotron losses - mainly non-thermal spectra
(Lorimer & Kramer 2006, Cambridge Press)
Phase
Nanda Rea CSIC-IEEC
The Pulsing Sky
Pulses at 1/10th true rate
Radio pulsars: Fermi results!
X-ray binaries: accretion and thermonuclear bursts
- Accretion physics: strong outburst and thermonuclear reactions - The accretion accelerates the neutron star rotation à ending up as millisecond pulsars
Nanda Rea CSIC-IEEC
Binary pulsars: test of General Relativity and NS EoS
- Same radio pulsar physics - Strong gravitational waves emitters - Best test of General Relativity: right at 99.98 % - Test NS equation of state
(Burgay et al. 2003, Nature; Lyne et al. 2004, Science) Nanda Rea CSIC-IEEC
1- Radio pulsars: ~2000
4- X-ray Dim Isolated Neutron Stars (XINSs): ~9
5- Rotating Radio Transients (RRATs): ~70
3- X-ray binary neutron stars: ~800
7- Magnetars (AXPs & SGRs): ~20
Different classes of neutron stars
2- Binary pulsars: ~60
Nanda Rea CSIC-IEEC
6- Central Compact Objects (CCOs): ~5
Thermally emitting neutron stars: ~ pure blackbodies
Nearby, old neutron stars, on the final phase of their cooling. Hotter than normal pulsars probably because of their high B-fields
(van Kerkwijk & Kaplan 2007, Ap&SS)
Nanda Rea CSIC-IEEC
(McLaughlin et al. 2006, N
ature)
(McLaughlin, R
ea et al., ApJ; E
SA P
ress Release)
- The most prolific burster and the most magnetic (B~7x1013 G), RRAT J1819-1458, has been recently discovered to be also a “normal” X-ray pulsar, and shows broad spectral features in its thermal X-ray spectrum.
Rotating Radio Transients (RRATs)
Nanda Rea CSIC-IEEC
Central Compact Objects (CCOs)
Nanda Rea CSIC-IEEC
- Young, thermally emitting neutron stars - Low magnetic fields ~1011 G - Buried by accretion? Or born such way?
1- Radio pulsars: ~2000
4- X-ray Dim Isolated Neutron Stars (XINSs): ~9
5- Rotating Radio Transients (RRATs): ~70
3- X-ray binary neutron stars: ~800
7- Magnetars (AXPs & SGRs): ~20
Different classes of neutron stars
2- Binary pulsars: ~60
Nanda Rea CSIC-IEEC
6- Central Compact Objects (CCOs): ~5
Neutron star formation & Magnetars
Magnetars are neutron stars with the highest magnetic fields
Those huge fields are believed to form either via alpha-dynamo soon after birth or as fossil fields from a very magnetic progenitor
Nanda Rea CSIC-IEEC
How magnetars were/are discovered
Short x/gamma-ray bursts (initially though to be GRBs)
Bright X-ray pulsars with 0.5-10keV spectra modelled by a
thermal plus a non-thermal component
Soft Gamma Repeaters
Anomalous X-ray Pulsars
Nanda Rea CSIC-IEEC
Bright X-ray transients!
No more distinction between Anomalous X-ray Pulsars and Soft Gamma Repeaters, all showing all kind of magnetars-like activity
No more distinction between Anomalous X-ray Pulsars, Soft Gamma Repeaters, and transient magnetars: all showing all kind of magnetars-like activity.
Transients
Swift-XRT INTEGRAL
COMPTEL
Fermi-LAT
(see Woods & Thompson 2006, Mereghetti 2008, Rea & Esposito 2011 for a review)
Magnetars general properties
• bright X-ray pulsars Lx ~ 1033-1036 erg/s
• strong soft and hard X-ray emission • short X/gamma-ray flares and long outbursts
• rotating with periods of ~2-12s • period derivatives of ~10-13-10-11 s/s • magnetic fields of ~1014-1015 Gauss • glitches and timing noise • faint infrared/optical emission (K~20; sometimes pulsed and transient)
(Israel et al. 2010)
(Abdo et al. 2010)
• pulsed fractions ranging from ~2-80 %
Nanda Rea CSIC-IEEC
Short bursts • the most common • they last ~0.1s • peak ~1041 ergs/s • soft γ-rays thermal spectra
Intermediate bursts • they last 1-40 s • peak ~1041-1043 ergs/s • abrupt on-set • usually soft γ-rays thermal spectra
Giant Flares • their output of high energy is exceeded only by blazars and GRBs • peak energy > 3x1044 ergs/s • <1 s initial peak with a hard spectrum which rapidly become softer in the burst tail that can last > 500s, showing the NS spin pulsations, and quasi periodic oscillations (QPOs)
(Israel et al. 2008)
(Kaspi et al. 2003)
Magnetar flares
Nanda Rea CSIC-IEEC
(Palmer et al. 2005)
(updated from Rea & Esposito 2011)
Outbursts • lasting between a few months to several years • peak ~1036 ergs/s • spectral variability: cooling
The Earth responding to magnetar flares
Nanda Rea CSIC-IEEC
(Mandea & Balasis 2006, Geophysical Journal)
(Palmer et al. 2005)
Why magnetars?
Can they be rotational powered as normal pulsars?
NO. X-ray emission overtaking their rotational budget.
Nanda Rea CSIC-IEEC
Why magnetars?
(Mereghetti, Israel & Stella 1998)
Can they be accretion powered by a low-mass companion star?
NO. Very stringent limits on possible companions.
Nanda Rea CSIC-IEEC
Another energy resevoir was needed
Why magnetars?
Nanda Rea CSIC-IEEC
• Magnetars have magnetic fields twisted up, inside and outside the star.
• The surface of a young magnetar is so hot that it
glows brightly in X-rays.
• Magnetar magnetospheres are filled by charged particles trapped in the twisted field lines, interacting with the surface thermal emission through resonant cyclotron scattering.
(Thompson & Duncan 1992; 1993; 1995;1996)
How magnetar persistent emission is believed to work?
!
"RCS ~RL
re
#
$ %
&
' ( "T ~ 10
5"T
!
RL ~ 8RNSBNS
Bcrit
"
# $
%
& '
1/ 31keV!(B
"
# $
%
& '
1/ 3
(Thompson, Lyutikov & Kulkarni 2002; Fernandez & Thompson 2008;
Nobili, Turolla & Zane 2008a,b; Rea et al. 2008, Zane et al. 2009)
Nanda Rea CSIC-IEEC
How magnetar outbursts and flares are believed to work?
• The twisted magnetic geometry of a magnetar, at intervals, it can twist up, and stresses build up in the neutron star crust causing outbursts, glitches, flares, and all sort of instabilities.
(Thompson, Lyutikov & Kulkarni 2002, ApJ 574, 332)
(Pons & Rea 2012, ApJ Letters, 750, L6)
Nanda Rea CSIC-IEEC
Varying injected energy
~1044 erg ~1043 erg
~1042 erg
~1041 erg
Varying initial quiescent luminosity
~1044 erg
25
What we believed until recently...
Nanda Rea CSIC-IEEC
1. A magnetar has necessarily a high dipole field !
2. Normal pulsars and
magnetars are two distinct classes of neutron stars.
!"# # #!
#!!$#!!%!"!#
!"#
##!
#!!#!!!
#!$
&'()*+,-'()./0).',1#!!#%,2,2!#3
&'()*+,123
!"#$%$&'(')*+
*
Critical Electron Quantum B-field
26
What we believed until recently...
Nanda Rea CSIC-IEEC
These turned out not to be totally true….
!"# # #!
#!!$#!!%!"!#
!"#
##!
#!!#!!!
#!$
&'()*+,-'()./0).',1#!!#%,2,2!#3
&'()*+,123
!"#$%$&'(')*+
*
1. A magnetar has necessarily a high dipole field !
2. Normal pulsars and
magnetars are two distinct classes of neutron stars.
Critical Electron Quantum B-field
1. Magnetars can be radio pulsars during outbursts
XTE 1810-197:showed radio pulsed emission during its outburst... the discovery of two other “radio-pulsar” magnetars followed soon...
(Camilo et al. 2006, Nature 442, 892; Camilo et al. 2007, ApJ 666, L93; Levin et al. 2010, ApJ 721, L33)
Nanda Rea CSIC-IEEC
(Rea, Pons, Torres, Turolla 2012, ApJ Letters, 748, L12, and highlighted in Science as Editors’ choice)
Nanda Rea CSIC-IEEC
1. Magnetars can be radio pulsars during outbursts
• P = 0.3 s
• B = 5x1013 Gauss
• Lspin-down = 200Lx~ 8x1036 erg/s
2. A “normal” X-ray pulsar showed magnetar activity...
PSR1846-0258: an energetic allegedly rotation-powered pulsar (with a high-B though...) showed SGR-like bursts
(Levin et al. 2010)
quiescence outburst
(Gavriil et al. 2008, Science, 319, 1802; Kumar & Safi-Harb, ApJ 678, L43)
Nanda Rea CSIC-IEEC
3. A magnetar was discovered having a low B-field...
SGR 0418+5729: discovered as a typical transient magnetar...
(van der Horst et al. 2010, ApJ 711, L1) (Esposito et al. 2010, MNRAS 405, 1787;
Rea et al. 2010, Science, 330, 944)
Nanda Rea CSIC-IEEC
Lx~1034 erg/s
Lx~1031 erg/s
kT (k
eV)
Flux
(erg
/s/c
m2 )
Time (days)
Time (s)
Cou
nt R
ate
(cou
nts/
s)
Magnetic field was: B < 7.5x1012 G
(Rea et al. 2010, Science, 330, 944)
Nanda Rea CSIC-IEEC
SGR 0418+5729: but having a low magnetic field!
P = 9.1s P < 6x10-15 s/s .
3. A magnetar was discovered having a low B-field...
SGR 0418+5729 magnetic field is: B ~ 7.1x1012 G
...now we have a possible B-field measurement for SGR 0418+5729, and a new low-B magnetar (Swift 1822.3-1606)!
(Rea et al. 2012, in prep)
Nanda Rea CSIC-IEEC
(Rea et al. 2012, ApJ, 754, 26; see also Sholtz et al. 2012)
3. A magnetar was discovered having a low B-field...
!"# # #!
#!!$#!!%!"!#
!"#
##!
#!!#!!!
#!$
&'()*+,-'()./0).',1#!!#%,2,2!#3
&'()*+,123
!"#$%$&'(')*+
*
Swift 1822.3-1606
SGR 0418+5729
(Rea et al. 2010, Science, 330, 944; Turolla et al. 2012, ApJ 740, 105) Nanda Rea CSIC-IEEC
Is this still compatible with the magnetar model?
Yes! Assuming that the crustal toroidal component of the B-field can be >> larger than the dipolar B-field we are measuring.
!"# # #!
#!!$#!!%!"!#
!"#
##!
#!!#!!!
#!$
&'()*+,-'()./0).',1#!!#%,2,2!#3
&'()*+,123
!"#$%$&'(')*+
*
Swift 1822.3-1606
SGR 0418+5729
Magnetars can be hidden inside many apparently quiet pulsars!
Which are the broader consequences of these discoveries?
** SN explosions A large number of strong-B neutron stars call for a key ingredient of the NS formation model: an extreme internal B should then be a common place rather than an exception
** GW radiation from newly born magnetars The GW background radiation produced by the formation of highly magnetic neutron stars is probably underestimated given the recent results.
** Gamma-ray bursts If a large fraction of the formed neutron stars have a strong B-field, hence GRBs due to the formation of such stars are way more frequent than predicted.
** Massive Stars If strong-B neutron stars are formed by the explosion of highly magnetic stars, there should be many more of such stars than predicted thus far
Nanda Rea CSIC-IEEC
Conclusions
Nanda Rea CSIC-IEEC
Magnetars are intriguing objects, and unique laboratories to test our knowledge on the physics of matter under extreme gravitational and magnetic fields.
We finally understood that behind the powerful magnetar emission there is not just a magnetic strength, but there are other important parameters: i.e. field geometry and evolution.
Many normal pulsars might be hiding a magnetar inside, hence magnetars might be the “typical” neutron stars rather than an exception, with all the due consequences.
The neutron star class appears as a very diverse class, with a variety of emission mechanisms. A unified scenario is still missing.