Transcript Non-Thermal Radio Emission from Binary Systems

Non-thermal radio emission
from binary systems
Josep M. Paredes
Exploring the Non-thermal Universe with Gamma Rays
On the occasion of the 60th birthday of Felix Aharonian
Barcelona, November 6 – 9, 2012
1
OUTLINE
1. Stellar radio sources
2. Microquasars
3. Non-accreting pulsars
4. New cases
5. Cataclysmic binaries
6. Colliding wind binaries
2
The radio flux density Sν at a frequency ν of a star measured at the Earth can
be expressed in terms of the Rayleigh-Jeans approximation to the BB formula
k: Boltzmann’s constant
TB: brightness temperature
ΩS: solid angle subtended by the star
Sν = 2 kν2 TB ΩS
c2
The maximum distance at which a star like the Sun would be observable at
5 GHz with a radio telescope with a typical minimum flux density of 1 mJy:
the quiet Sun:
the slowly varying component:
the very strong solar radio bursts:
0.2 pc
0.3 pc
7.6 pc
(Proxima Centauri is 1.3 pc away)
Solar-like emission would be seen from only a few nearby stars
Detectable radio stars have either
unusually large emitting surfaces
generally associated with stars
with massive stellar winds
or
enhanced brightness temperatures stars exhibiting powerful
non-thermal radio emission
3
Hertzprung-Russell diagram based on stellar radio detections
(Güdel 2002, Annu. Rev. Astron. Astrophys. 40, 217)
Radio emission has been detected
from all stages of stellar evolution
across the HR diagram
Most of the main sequence and
sub-giant objects are non-thermal
emitters
Most of the giants and many O-B
stars are thermal emitters (they
are big)
Novae and X-ray binaries not shown since the
active source is related to accretion onto a WD
and NS or BH respectively
4
Types of radio stars
Stars undergoing mass loss
OB and Wolf-Rayet stars: FF emission from an optically thick wind.
Non-thermal emission: relativistic e- embedded in the wind (Fermi acceleration in
shock fronts) or collision between the winds of two components of a binary
Be stars: FF emission as a consequence of the mass loss trough an equatorial disk
by centrifugal effects caused by the rapid rotation
Single red giants and supergiants: the winds are cool and only partially
ionised. Detectable if near
VV Cephei stars: cool supergiant + B V companion. FF emission from a subregion
of the supergiant wind ionized by UV photons from the hotter companion
Pre-main sequence stars: T Tauri stars. Classical (CTT): FF from ionised gas.
Weak-lined (WTT): non-thermal radio emission from magnetically active regions
5
Stars exhibiting enhanced solar activity
The sun exhibits non-thermal emission and flare activity associated with high-energy particles in the
chromosphere and corona. The same phenomenon on larger scales is characteristics of many cool
(primarily K-M) stars on and above MS. The energy supply is thought to be the release of stellar
magnetic field energy by reconnection of field lines. The fields may be generated by the dynamo
mechanism.
Flare stars: Single stars that exhibit intense flares from X-rays to radio waves. The
coronal gas is bound by magnetic loops (several 1000 G) covering most of the stellar
surface. Stellar flares due to coherent emission (at lower frequencies) and incoherent
gyrosynchrotron emission (higher frequencies). Prototype: UV Ceti, YZ CMi, AD Leo
Close binaries: RS Canum Venaticorum stars: Solar type + cool subgiant.
Gyrosynchrotron emission.
Magnetic activity is enhanced by the high rotation rate of the active subgiant which is
synchronised to the period of orbital revolution by tidal coupling. Particle acceleration
may also occur in the intrabinary plasma if the interaction between the fields of the two
stars produce reconnection.
Pre-main sequence stars: Some of the properties of WTT stars are similar to
the RS CVn stars. Gyrosynchrotron emission from starspot regions
6
Chemically peculiar stars
CP stars or Bp-Ap stars: MS stars characterised by over and under abundances of certain
chemical elements and by strong (1-10 kG) dipolar magnetic fields. No convection: B may
be a fossil remnant of dynamo fields generated in the PMS phase. Radio characteristics (flat
radio spectrum, circular polarisation, non-thermal TB) consistent with gyrosynchrotron
emission
Radio emission related to mass transfer in binaries
The transfer process may involve Roche lobe overflow or accretion from a stellar wind
from a normal star to a white WD, NS or BH.
Cataclysmic variables: Classical novae: semi-detached binaries (MS+WD).
Nova outbursts (intervals of ~ 104 yr) occurs when accreted H rich material accumulates
on the outer surface of the WD leading to a thermal runaway and explosive ejection of the
outer envelope. FF emission from the expanding ionised ejecta. Synchrotron radiation
from particles accelerated in a shock (within the ejecta or interaction between the nova
ejecta and a dense gas cloud). Magnetic cataclysmic variables (AM Her
stars): late-type MS+magnetic WD. The mass transfer process is modified by the
presence of strong magnetic fields (105-107 G). Highly variable flare-like non-thermal
radio emission. Symbiotic stars: interacting binaries (cool giant (read giant) + hot
companion). FF emission from a circumstellar ionised envelope.
X-ray binaries: x-ray emission produced by accretion. HMXB (130) and LMXB
(150). 20% detected in radio. Synchrotron radiation from relativistic particles
interacting with magnetic fields.
7
Binaries with synchrotron radio emission
Be/X-ray transients
HMXRB
X-ray binaries
SG/obscure binaries
BH binaries
LMXRB
1. Accreting binaries
Microquasars
Relativistic jet
Cataclysmic binaries
2. Non-accreting binaries
Colliding-wind massive binaries (OB+WR,
OB+OB,WR+WR)
Pulsar wind binaries (e.g. PSR B1259-63)
Sources of relativistic particles  HE and VHE gamma-rays
could be produced
2
2
4
2 B
 dE 
4
 dE 
  T cγ
    T c γmax Uphoton
 
max
2
 dt  I.C. 3
 dt Sync 3
8
Possible scenarios
• An accretion disk is formed by mass transfer.
Radio emission
Synchrotron Radiation
Microquasar
• Display bipolar jets of relativistic plasma.
• The jet electrons produce radiation by synchrotron
emission when interacting with magnetic fields.
• VHE emission is produced by inverse Compton
scattering when the jet particles collide with stellar
UV photons, or by hadronic processes when
accelerated protons collide with stellar wind ions.
[Mirabel & Rodríguez 1998, Nature 392, 673; Aharonian &
Atoyan 1998, NewAR 42, 579; Romero et al. 2003, A&A, 410,
L1; Bosch-Ramon et al. 2006, A&A, 447, 263]
Non-accreting pulsar
e- e
e- -
UV - Opt
Donor star
e
-
OB Star
X-ray
Disk black body or
Corona power-law
e-
e-
Γe~105
Gamma-ray
Inverse Compton Scattering
• The relativistic wind of a young (ms) pulsar is contained
by the stellar wind.
• Particle acceleration at the termination shock leads to
synchrotron and inverse Compton emission.
• After the termination shock, a nebula of accelerated
particles forms behind the pulsar.
• The cometary nebula is similar to the case of isolated
pulsars moving through the ISM.
[Maraschi & Treves 1981, MNRAS, 194, P1; Dubus 2006, A&A,
456, 801; Sierpowska-Bartosik & Torres 2007, ApJ, 671, L145]
9
From Moldón 2012
Source
System
Type
Orbital
Period (d)
Radio
Structure
(AU)
Radio
X-ray
GeV
PSR
B1259-63
O9.5Ve +
NS
1237
Cometary tail
~ 120
P
P
T
P
LS I +61 303
B0Ve + ?
26.5
Cometary
tail?
10 – 700
P
P
P
P
LS 5039
O6.5V((f
)) +?
3.9
Cometary
tail?
10 – 1000
P
P
P
HESS J0632+057
B0Vpe +
?
321
Elongated
(few data)
~ 60
V
P
?
P?
1FGL J1018.65856
O6.5V((f
)) +?
16.6
?
P
P
P
?
Cygnus X-1
O9.7I +
BH
5.6
Jet
40 + ring
persistent
P
T?
T?
Cygnus X-3
WR +
BH?
4.8h
Jet
Persistent
& burst
P
P
?
persistent
TeV
10
Microquasars
● HMXB, O9.71+BH
Cygnus X-1 Stellar Mass Black Hole
5 pc (8’) diameter ring-structure of bremsstrahlung emitting
ionized gas at the shock between (dark) jet and ISM.
VLA
WSRT
VLBA+VLA
Gallo et al. 2005, Nature
15 mas
30 AU
Stirling et al. 2001, MNRAS 327, 1273
Martí et al. 1996, A&A 306, 449
11
Cygnus X-3 Strong radio outbursts
Microquasars
● HMXB, WR+BH?
 Exhibits flaring to levels of 20 Jy or more
 Modelling Cyg X-3 radio outbursts:
particle injection into
twin jets Martí et al. 1992, A&A 258, 309
VLBA
VLA, 5 GHz
Martí et al. 2001, A&A 375, 476
12
Miller-Jones et al. 2004, ApJ 600, 368
Binary pulsar systems
PSR B1259-63 Young pulsar wind interacting with the companion star
The first variable galactic source of VHE
PSR B1259-63 / LS 2883: O8.5-9 Ve
(Negueruela et al. 2011, ApJL, 732, L11)
Aharonian et al. 2009, A&A 507,
389
 Dense equatorial circumstellar disk + 47.7 ms radio
pulsar, P = 3.4 yr, e = 0.87
 No radio pulses are observed when the NS is
behind the circumstellar disk (free-free absorption)
 Orbital plane of the pulsar inclined with respect to
the disk (Melatos et al. 1995, MNRAS 275, 381;
Chernyakova et al. 2006, MNRAS 367, 1201)
Abdo et al. 2011,
ApJ 736, L11
Chernyakova et al. 2009,
MNRAS 397, 2123
Johnston
1999
PSR B1259 / LS 2883
et
HESS June 2007
al.
PSR B1259-63. Nearly all the spin-down power is released in HE
gamma rays (Abdo et al. 2011). Doppler boosting suggested (Tam
et al. 2011), but very fine tuning is needed(!).
13
Binary pulsar systems
Australian Long Baseline Array (LBA)
2.3 GHz
Extended radio structure
Moldón et al. 2011, ApJ 732, L10
Total extension of the nebula: ~ 50 mas, or 120 ± 24 AU
The red crosses marks the region where the pulsar should be contained in each run
Kinematical model
Moldón et al. 2011, ApJ
732, L10
This is the first observational evidence that non-accreting pulsars orbiting massive
Shock between the relativistic pulsar
producestellar
variable
extended
wind stars
and acan
spherical
wind
(Dubus radio emission at AU scales
2006, A&A 456, 801)
The evolution of the nebular flow after the
shock is described in Kennel & Coroniti (1984)
PSR B1259 / LS 2883
14
Microquasar or pulsar scenario
?
LS I +61 303
HMXB, B0Ve+NS?
COS-B γ-ray source CG/2CG0.8-0.5
135+01
periodicity
Hermsen et al. 1977, Nature 269, 494
Radio (P=26.496 d) Taylor & Gregory 1982, ApJ 255, 210
Optical and IR Mendelson & Mazeh 1989, MNRAS 239, 733;
Paredes et al. 1994 A&A 288, 519
X-rays Paredes et al. 1997 A&A 320, L25; Torres et al. 2010, ApJ 719, L104
HE gamma-rays Abdo et al. 2009, ApJ 701, L123
VHE gamma-rays Albert et al. 2009, ApJ 693, 303
4.4 yr periodicity Radio (P= 1667 d)
Paredes 1987, PhD Thesis; Gregory 2002, ApJ 575, 427
0.5-0.8
Paredes et al. 1990, A&A 232, 377
Strickman et al. 1998, ApJ 497, 419
Gregory 2002, ApJ 575, 427
0.5-0.8
15
VLBA
Microquasar or pulsar scenario
?Jet-like features have been reported several times, but show a puzzling behavior (Massi
et al. 2001, 2004). VLBI observations show
Microquasars Workshop, Como, Setember 2006)
G
H
a rotating jet-like structure (Dhawan et al. 2006, VI
F
E
A
I
D
B
C
J
NOT TO SCALE
Observer
3.6cm images, ~3d apart, beam 1.5x1.1mas or 3x2.2 AU.
Semi-major axis: 0.5 AU
16
LS I + 61 303
Microquasar or pulsar scenario
?
Repeatability of the LS I +61 303 morphology
Highly stable morphology probably indicates interaction of steady outflows
Striking similarity
8 GHz , 2006 July
orbital phase 0.62
5 GHz , 2006 October 25 (contours)
8 GHz , 2006 February 2 (gray scale)
Dhawan et al. (2006)
8 GHz , 2007 September
(Moldón 2012, PhD thesis)
(Albert et al. 2008)
LS I + 61 303
2006
2007
17
Microquasar or pulsar scenario
?
LS 5039 HMXB,
Orbital morphological variability
O6.5V+NS?
 Images at the same phase
have similar morphology
 Periodic orbital modulation
of the radio morphology of
LS 5039, stable over several
years
 Changes in the radio
structure are compatible with
the presence of a young non
accreting NS
Moldón et al. 2012, A&A
arXiv 1209.6073
18
LS 5039
VLBA image
Model
LS 5039
Moldón et al. 2012, A&A
arXiv 1209.6073
Model of the VLBI structure of LS 5039
considering an outflow starting at 10a and
using a similar bending as in BoschRamon et al. 2012. Includes synchrotron
and adiabatic losses.
19
On the lack of pulsations in LS 5039 and LS I +61 303
Courtesy of Javier Moldón
LS 5039
Porb = 3.9 day
O6.5 V + ?
Radio pulses
disappear ?
LS I +61 303
Porb = 26.5 day
B0Ve + ?
Radio pulses
disappear ?
PSR B125963
Porb = 3.4 yr
O8.5Ve +
pulsar
Radio pulses
disappear
20
HESS J0632+057
New cases
Skilton et al. 2009, MNRAS 399, 317
Hinton et al. 2009, ApJ 690, L101
 Be star MWC 148
 XMM-Newton
Δ1RXS J063258.3+054857
Swift X-ray periodicity: P=321 ± 5 days  strong evidence for
binary nature Bongiorno et al., 2011, ApJ 737, 11
confirmed by Casares et al., 2012, MNRAS 421, 1103
In Feb. 2011 Swift reported increased X-ray activity (Falcone et al. 2011, Atel # 315
VERITAS and MAGIC detected elevated TeV gamma-ray emission (Ong et al. 2011, Atel # 3153; Mariotti
et al. 2011, Atel # 3161)
VLBI counterpart
(Moldón et al. 2011, Atel # 3180)
Confirmed association with Be star
 Confirmed the non-thermal nature
of
the radio source
 Discovery of extended emission

HESS J0632+057
Moldón et al.
2011, A&A 533, L7
21
1FGL J1018.6-5856
New cases
Ackermann et al. 2012, Fermi Col., Science 335, 189
• 1FGL J1018.6-5856 is one of the brighter Fermi sources
Fermi
• LAT spectrum similar to a pulsar - but no pulsations seen
• Optical counterpart ~O6V((f)), just like LS 5039
Flux modulated with a 16.6 d period
Swift-XRT
 X-ray flare-like behaviour near phase 0,
coinciding with gamma-ray maximum
 An spatially coincident variable radio
source
ATCA: 5.5 GHz, 9 GHz
 Radio structure ?
22
V407 Cygni
Cataclysmic binaries
Abdo et al. 2010, Science 329, 817
Mira-type pulsating red giant + WD (symbiotic binary)
•The first nova explosion ever detected in grays
•The gamma-ray emission has been attributed
to particles accelerated by the interaction of the
nova shock with the dense Mira wind
JVLA
Chomiuk et al. 2012, ApJ, in press
arXiv 1210.6029
Radio light curve dominated by the wind of the
Mira giant companion
Not observed a thermal signature from the nova ejecta or synchrotron emission from
the shock, due to the fact that these components were hidden behind the absorbing
screen of the Mira wind.
23
Colliding winds in massive binary systems
Collision of supersonic winds in massive star binaries (WR
have very strong winds, v∞ up to 5000 km s-1)  strong
shocks where both e and p can be efficiently accelerated up to
relativistic energies through first-order Fermi mechanism
(Eichler& Usov 1993)
Non-thermal synchrotron emission from the colliding wind
region in WR140 has been detected (Dougherty et al. 2005,
ApJ 623, 447)
 the presence of highly relativistic electrons.
These systems may also be embedded in dense photon fields
where IC losses would be unavoidable,
 making CWBs potential high-energy emitters
WR 140
AGILE (Tavani et al. 2009) and Fermi-LAT (Abdo et al. 2010) established high-energy γ-ray emission in
spatial coincidence with the location of η Car.
A weak but regular Fermi flux decrease over time has been detected (Reitberger et al. 2012) and
interpreted in a colliding-wind binary scenario for orbital modulation of the γ-ray emission.
 First unambiguous detection of GeV γ-ray emission frrom a colliding-wind massive star
24
Summary
The MQs and Binary pulsar systems with HE and/or VHE gamma-ray emission






Are (synchrotron) radio emitters
Periodic at all wavelengths (?)
Have a bright companion (O or B star)  source of seed photons
for the IC emission
MQs: Jet radio structures
Binary (non-accreting) pulsar systems orbiting massive stars
– can produce variable extended radio emission at AU scales
– repeatability of their radio structures with the orbit of the binary
system
(PSR B1259−63, LS 5039, LS I +61 303)
VLBI radio observations are a common link, useful to understand the
behavior of gamma-ray binaries. Can put constrains on physical
parameters of the system.
25