150
A. Patruno and A. L. Watts
about the presence of pulsations, and the accretion-powered origin appears, so far,
the most promising explanation [44]. The case of SAX J1748.9-2021 is slightly
different: pulsations were detected sporadically in several data segments and in three
(2001, 2005 and 2009–2010 [2, 253, 261]) out of four outbursts observed (the first
being in 1998). It is unclear why these three systems show pulsations intermittently.
In HETE J1900.1-2455, an increase in pulse fractional amplitude was reported
approximately in coincidence with the occurrence of Type I X-ray bursts [101],
followed by a steady decrease. On other occasions the pulsations appeared a few
hours before or after a burst, indicating that pulsations might be linked somehow
with some yet to be identified property of the NS envelope. In SAX J1748.9-2021
the pulsed amplitudes showed some abrupt changes in amplitude and/or phase in
coincidence with about 30% of the observed Type I X-ray bursts [2, 253]. The
pulsations, however, displayed a more diverse behaviour than in HETE J1900.12455, without the typical steady decrease of fractional amplitudes. A period of
global surface activity during which both Type I bursts and pulsations are produced
might be at the origin of this link [253]. The single pulsating episode of Aql X-1 had
instead no clear connection with Type I bursts, even though Aql X-1 is a bursting
LMXB.
The spin frequency derivative of HETE J1900.1-2455 was measured over a
baseline of 2.5 years, an unprecedented long baseline for an AMXP (whose
outbursts last usually less than 100 days). This has indirectly provided hints on the
physical origin of such a period of global surface activity in intermittent sources. The
spin frequency derivative exhibited an exponential decay in time that was interpreted
as evidence of the screening of the NS magnetic field [250]. It was proposed that
intermittency originates because the magnetic field strength drops by almost three
orders of magnitude on a timescale of few hundred days so that the disk cannot be
truncated and only a (very) shallow layer of gas can be channeled to form (weak)
pulsations (see also Sect. 4.6.3 for a discussion of the model).
One feature that intermittent pulsars share is that the long term average mass
accretion rate ˙
M is higher than for the persistent AMXPs and smaller than for the
bright non-pulsating systems (like Sco X-1 and the other Z sources). Calculating
the precise value of ˙
M is rather difficult, since it depends on poorly constrained
parameters in most LMXBs, like the distance d, the X-ray to bolometric flux
conversion and the recurrence time of the outburst. However, even considering
these caveats, it seems clear that at least the brightest systems have never showed
pulsations. Whether it is the high mass accretion rate that determines the lack of
pulsations or some other properties shared by the brightest systems is still unclear.
See Sect. 4.6 for an extended discussion on the mechanism that might prevent the
formation of pulsations in LMXBs.
A. Patruno and A. L. Watts
about the presence of pulsations, and the accretion-powered origin appears, so far,
the most promising explanation [44]. The case of SAX J1748.9-2021 is slightly
different: pulsations were detected sporadically in several data segments and in three
(2001, 2005 and 2009–2010 [2, 253, 261]) out of four outbursts observed (the first
being in 1998). It is unclear why these three systems show pulsations intermittently.
In HETE J1900.1-2455, an increase in pulse fractional amplitude was reported
approximately in coincidence with the occurrence of Type I X-ray bursts [101],
followed by a steady decrease. On other occasions the pulsations appeared a few
hours before or after a burst, indicating that pulsations might be linked somehow
with some yet to be identified property of the NS envelope. In SAX J1748.9-2021
the pulsed amplitudes showed some abrupt changes in amplitude and/or phase in
coincidence with about 30% of the observed Type I X-ray bursts [2, 253]. The
pulsations, however, displayed a more diverse behaviour than in HETE J1900.12455, without the typical steady decrease of fractional amplitudes. A period of
global surface activity during which both Type I bursts and pulsations are produced
might be at the origin of this link [253]. The single pulsating episode of Aql X-1 had
instead no clear connection with Type I bursts, even though Aql X-1 is a bursting
LMXB.
The spin frequency derivative of HETE J1900.1-2455 was measured over a
baseline of 2.5 years, an unprecedented long baseline for an AMXP (whose
outbursts last usually less than 100 days). This has indirectly provided hints on the
physical origin of such a period of global surface activity in intermittent sources. The
spin frequency derivative exhibited an exponential decay in time that was interpreted
as evidence of the screening of the NS magnetic field [250]. It was proposed that
intermittency originates because the magnetic field strength drops by almost three
orders of magnitude on a timescale of few hundred days so that the disk cannot be
truncated and only a (very) shallow layer of gas can be channeled to form (weak)
pulsations (see also Sect. 4.6.3 for a discussion of the model).
One feature that intermittent pulsars share is that the long term average mass
accretion rate ˙
M is higher than for the persistent AMXPs and smaller than for the
bright non-pulsating systems (like Sco X-1 and the other Z sources). Calculating
the precise value of ˙
M is rather difficult, since it depends on poorly constrained
parameters in most LMXBs, like the distance d, the X-ray to bolometric flux
conversion and the recurrence time of the outburst. However, even considering
these caveats, it seems clear that at least the brightest systems have never showed
pulsations. Whether it is the high mass accretion rate that determines the lack of
pulsations or some other properties shared by the brightest systems is still unclear.
See Sect. 4.6 for an extended discussion on the mechanism that might prevent the
formation of pulsations in LMXBs.