3 Magnetars: A Short Review and Some Sparse Considerations
125
period longer than 1000 s, the neutron star dipole field must be 10 14 G [42].
This is unless the accretion rate is very low, but it would be orders of magnitude
below the rate necessary to account for the luminosity of these persistent Be X-ray
binary systems, hence the puzzle. However, in the recent model of quasi-spherical
settling accretion in wind-fed high-mass X-ray binaries by Shakura et al. [195], the
equilibrium period can be of ∼1000 s even for ‘ordinary’ magnetic fields of ∼ 10 12 –
10 13 G and acceptable accretion rates. The model has been applied successfully
in populations studies and to model samples and singles sources (e.g. [30, 143]),
including the slowest known accreting pulsar, in the high-mass X-ray binary system
AX J1910.7+0917 (spin period of 36.2 ks; [197]).
There is however an exception, the Be X-ray binary SXP 1062 in the Small
Magellanic Cloud, which has a spin period of 1062 s and is robustly associated
to a supernova remnant with kinematic age of (2–4) × 10 4 yr (Hénault-Brunet et
al. [100]; Haberl et al. [97] propose an even younger age using a temperature–size
relationship). In fact, for typical values of magnetic field and accretion rate, it would
have been impossible for the neutron star to enter the propeller stage and become
an accretor within the time constraint of a few ×10 4 yr dictated by the age of the
supernova remnant. After considering several possible scenarios, [180] proposed a
neutron star born with an initial magnetic field of 10 14 G that then decayed to a
present value of ∼10 13 G (derived in the assumption that the star is rotating close to
the equilibrium period).
Magnetars are increasingly popular also in the field of ultraluminous X-ray
sources (ULXs). These sources—a few hundreds of them are known—are observed
in off-nucleus regions of nearby galaxies at X-ray luminosities exceeding a few
10 39 erg s −1 [123]. Since this threshold for isotropic luminosity is larger than the
Eddington limit for spherical accretion of fully ionized hydrogen onto a ∼10M compact object (a scale value of the black holes of stellar origin observed in
our Galaxy), ULXs were considered the observational manifestation of massive
black holes of stellar origin (80–100 M ) and, the brightest ones in particular,
promising candidates of intermediate-mass black holes of 10 3 –10 5 M [156]. For
these reason, the recent discovery in three ULXs (M82 X–2, NGC 5907 ULX, and
NGC 7793 P13) of pulsars with spin periods from 0.4 to 1.4 s has been a blow,
showing both that some ULXs (even in the high side of their luminosity distribution)
may host neutron stars and that neutron stars can achieve extreme supper-Eddington
luminosities [10, 121, 122]. The most luminous of the bunch, NGC 5907 ULX,
which has a period of ∼1.1 s, was observed at a maximum X-ray luminosity of
∼ 10 41 erg s −1 , more than 500 times the Eddington limit for a 1.4-M neutron star
[121]. In principle, a neutron star with a magnetic field of 10 15 G could attain
such a super-Eddington luminosity [40, 160], since the magnetic field reduces the
electron scattering cross section (see also Sect. 3.2.2.1). However, this explanation
is not viable in the cases of NGC 5907 ULX and NGC 7793 P13, because such a
huge magnetic field coupled with the rapid spinning of the star would inhibit the
accretion via the propeller mechanisms. A possible solution proposed by Israel et
al. [121] is that the magnetic field is indeed of a few 10 14 G at the neutron star
125
period longer than 1000 s, the neutron star dipole field must be 10 14 G [42].
This is unless the accretion rate is very low, but it would be orders of magnitude
below the rate necessary to account for the luminosity of these persistent Be X-ray
binary systems, hence the puzzle. However, in the recent model of quasi-spherical
settling accretion in wind-fed high-mass X-ray binaries by Shakura et al. [195], the
equilibrium period can be of ∼1000 s even for ‘ordinary’ magnetic fields of ∼ 10 12 –
10 13 G and acceptable accretion rates. The model has been applied successfully
in populations studies and to model samples and singles sources (e.g. [30, 143]),
including the slowest known accreting pulsar, in the high-mass X-ray binary system
AX J1910.7+0917 (spin period of 36.2 ks; [197]).
There is however an exception, the Be X-ray binary SXP 1062 in the Small
Magellanic Cloud, which has a spin period of 1062 s and is robustly associated
to a supernova remnant with kinematic age of (2–4) × 10 4 yr (Hénault-Brunet et
al. [100]; Haberl et al. [97] propose an even younger age using a temperature–size
relationship). In fact, for typical values of magnetic field and accretion rate, it would
have been impossible for the neutron star to enter the propeller stage and become
an accretor within the time constraint of a few ×10 4 yr dictated by the age of the
supernova remnant. After considering several possible scenarios, [180] proposed a
neutron star born with an initial magnetic field of 10 14 G that then decayed to a
present value of ∼10 13 G (derived in the assumption that the star is rotating close to
the equilibrium period).
Magnetars are increasingly popular also in the field of ultraluminous X-ray
sources (ULXs). These sources—a few hundreds of them are known—are observed
in off-nucleus regions of nearby galaxies at X-ray luminosities exceeding a few
10 39 erg s −1 [123]. Since this threshold for isotropic luminosity is larger than the
Eddington limit for spherical accretion of fully ionized hydrogen onto a ∼10M compact object (a scale value of the black holes of stellar origin observed in
our Galaxy), ULXs were considered the observational manifestation of massive
black holes of stellar origin (80–100 M ) and, the brightest ones in particular,
promising candidates of intermediate-mass black holes of 10 3 –10 5 M [156]. For
these reason, the recent discovery in three ULXs (M82 X–2, NGC 5907 ULX, and
NGC 7793 P13) of pulsars with spin periods from 0.4 to 1.4 s has been a blow,
showing both that some ULXs (even in the high side of their luminosity distribution)
may host neutron stars and that neutron stars can achieve extreme supper-Eddington
luminosities [10, 121, 122]. The most luminous of the bunch, NGC 5907 ULX,
which has a period of ∼1.1 s, was observed at a maximum X-ray luminosity of
∼ 10 41 erg s −1 , more than 500 times the Eddington limit for a 1.4-M neutron star
[121]. In principle, a neutron star with a magnetic field of 10 15 G could attain
such a super-Eddington luminosity [40, 160], since the magnetic field reduces the
electron scattering cross section (see also Sect. 3.2.2.1). However, this explanation
is not viable in the cases of NGC 5907 ULX and NGC 7793 P13, because such a
huge magnetic field coupled with the rapid spinning of the star would inhibit the
accretion via the propeller mechanisms. A possible solution proposed by Israel et
al. [121] is that the magnetic field is indeed of a few 10 14 G at the neutron star