Transcript Optical observations of Quiescent Low Mass X

Binary millisecond X-ray pulsars
Paul Callanan and Mark Reynolds
Department of Physics
University College Cork
Alexei Filippenko,
Department of Astronomy, Berkeley
Peter Garnavich,
Department of Physics,
University of Notre Dame.
Galactic X-ray source distribution:
QuickTime™ and a
TIFF (LZW) decompressor
are needed to see this picture.
LMXBs (open circles) and HMXBs (filled circles)
(Grimm et al 2002 from RXTE ASM)
• More than 300 X-ray binaries known, both persistently bright
and quiescent.
• Galactic X-ray emission dominated by the Low Mass systems in
the bulge (M2<1-2 Mo)
• High mass systems in the disk/spiral arms
Grimm et al 2002
• Globular cluster sources - 13 LMXB, both persistently
bright/transient (Pooley et al 2002): fainter sources include
quiescent LMXBs, CVs, millisecond pulsars, magnetically active
binaries (+ …: Verbunt and Lewin, 2005)
• Galactic Centre/diffuse emission (e.g. Muno et al 2003, Hands et
al 2004) - faint XRN, HMXBs/magnetic CVs, diffuse emission
High Mass X-ray binaries
Vast majority(~96) pulsate (periods
ranging from 69 msec to 1400 s).
QuickTime™ and a
TIFF (LZW) decompressor
are needed to see this picture.
Recent neutron star mass
determinations for HMXBs…
Van der Meer et al 2005
QuickTime™ and a
TIFF (LZW) decompressor
are needed to see this picture.
Low Mass X-ray binaries
• More than half of all known Galactic XRB are LMXB
• Orbital periods 11 mins to 16.5 days.
• Lx (persistently bright) ~1036 -1038 ergs/s. Lower Lx sources
often exhibit X-ray bursts.
X-ray Novae (aka Soft X-ray Transients)
• ~half of all known LMXBs are transient. Typical recurrence
times are thought to be ~10-50 years.
• Lx(quiescent) as low as 1030 ergs/s: Lx(max) >~ 1039 ergs/s
in some systems.
• Orbital periods ~4 hrs - 33.5 days
• Of the 18 binaries thought to contain a black hole (ie for
which Mx>3 Mo from radial velocity studies), 15 are XRN.
System parameters for black hole XRN
QuickTime™ and a
TIFF (LZW) decompressor
are needed to see this picture.
(Orosz)
Compact object
mass estimates
(but beware
- e.g. Reynolds
et al 2006)
QuickTime™ and a
TIFF (LZW) decompressor
are needed to see this picture.
Accreting binary millisecond pulsars
• 7 discoveries since the first (in 1996), SAX J1808.4-3658
(bursting, pulsating).
• Orbital periods from 40 mins ->4.3 hrs:
spin periods from 1.67-5.4 msec: Lmax ~1036 ergs/s
• Dramatic confirmation of the link between accreting LMXBs and
millisecond pulsars.
• Evidence for irradiation of secondary by pulsar spin down flux in
at least one system.
• Spin periods now known for ~20 LMXBs (these + burst oscillations
+ persistently bright pulsators).
• In theory, ideal for constraining neutron star mass…
IGR J00291+5934 - discovered by INTEGRAL during a routine
Galactic Plane Scan on the 2nd of December 2004.
The optical counterpart was subsequently identified two days later and
was found to have a magnitude R ≈ 17.4 (Fox et al., 2004). Lx~1036
ergs/s.
We obtained a single 300s
LRIS spectrum of the
proposed optical counterpart,
which we display in Figure 6,
was obtained 10 days post
outburst(Filippenko et al.,
2004). We observe broad
(FWHM = 1200 km/s)
emission lines of Hα 656nm
(EW = 0.96 nm), Hβ 486nm
(EW = 0.54 nm), and HeI
667.8 nm (EW = 0.1 nm), as
well as narrow
(FWHM = 300 km/s), very
weak (EW = 0.06 nm) HeII
468.6 nm emission.
Outburst
photometry:
variability, but
nothing on the
orbital period.
WIYN (4m)
IGR J00291+5934 is the 6th member of the class of accretion
powered millisecond X-ray pulsars to be discovered and, with an
orbital period of 2.45 hours, the third ‘long’ orbital period system.
The other long period sources are SAX J1808.4-3658, (Wijnands et
al., 1998; Chakrabarty et al., 1998), XTE J1814-338 (Markwardt et
al., 2003b) and HETE J1900.1-2455 (Vanderspek et al., 2005) with
periods of 2.01 hr, 4.28 hr and 1.39 hr respectively.
The final 3 members of the class are ultra-compact binaries with
periods of ∼40 minutes (XTE J1751-305, Markwardt et al., 2002:
XTE J0929-314, Galloway et al., 2002 and XTE J1807-294,
Markwardt et al., 2003a). Even though IGR J00291+5934 and SAX
J1808.4-3658 are geometrically similar, the magnitude of the
former in quiescence is at least 2 magnitudes fainter; R ≈ 23
compared to 20.9 ±0.1 (Homer et al., 2001).
As the distance to each of these systems is comparable (Jonker et
al. 2005; in’t Zand et al. 2001), this implies that IGR J00291+5934
is intrinsically, at least 2 magnitudes fainter than SAX J1808.43658 in quiescence. This may suggest that the spindown luminosity
of the pulsar in IGR J00291+5934 is considerably less than that
of SAX J1808.4-3658, resulting in a fainter quiescent
counterpart. However more accurate distance and reddening
estimates are required to confirm this.
Quiescent observations
Can we see the secondary ?
… we might expect, by analogy with Cataclysmic Variables…
IGR J00291+5934: 30 min Keck integration.
Data with NextGen model (T=2845 K).
WIYN quiescent photometry + ellipsoidal fit
i = 60 degrees, T=4000 K
Unfortunately, this is not the whole story….
So…
Continuum from the disk absent - not only because of
short orbital period ? “Naked” secondary - Mv~14.4,
V-I~3.5, consistent with the data.
Could be at relatively high inclination, but we need to
reconcile photometric and spectroscopic fits.
No evidence for blue excess (in contrast to SAX
J1808.4-3658) - ie for shockfront between wind from
secondary and neutron star.
Radial velocity measurements/mass estimates tricky…
Much remains to be learned
by comparing these systems
to the short period CVs…
Orbital period 81 mins