124
P. Esposito et al.
3.2.6 Magnetars in Binary Systems
It is not clear whether magnetar exist in binary systems but, although it is possible
that magnetars to form have to sacrifice the companion, binarity may be an
important ingredient for the production of magnetars for at least two reasons.
Firstly, if magnetars descend from particularly massive stars (20 M ), the binary
interaction may prevent the formation of a black hole instead of a neutron star.
Secondly, in the case the magnetic fields of magnetars are formed through dynamo
amplification, even if magnetars form from stars with lower masses, a binary system
may help the stellar core to maintain the angular momentum necessary for the
dynamo mechanisms. On the other hand, in the fossil field scenario, in which
the neutron star magnetic field reflects that of the stellar precursor, progenitors
belonging to binary systems may be disfavoured, since magnetic hot stars in
compact binary systems are very rare [161].
Both scenarios for the origin of magnetar magnetic field require a very massive
progenitor, heavier than ∼20 M . For the fossil field, because there seems to be in
main sequence stars a trend of stronger magnetic fields with larger masses (Ferrario
and Wickramasinghe [70] and references therein). In the dynamo hypothesis,
because to produce neutron stars rotating fast enough to generate a magnetar field
via α-ω dynamo, a star more massive than ∼20–35 M star is necessary [98].
There are also observational facts that favour very massive stars as the magnetar
precursors. Considering the spatial distribution of the known magnetars, it has been
noted that their height on the Galactic plane is smaller than that of OB stars. This
suggests that they were produced by the most massive O stars [164]. Furthermore,
there are some tentative associations of magnetars with massive star clusters.
In the most convincing case, CXOU J164710.2–455216 in Westerlund 1, the
young age of the cluster (∼4 Myr) implies a progenitor with minimum mass of
∼40 M [158]. 3 To allow such a massive star to produce a neutron star, Clark et al.
[33] suggested a binary with (41 + 35) M stars and an orbital period shorter than
8 days. 4 The idea is that the binary interaction drives the primary in a Wolf–Rayet
star that through its powerful stellar winds loses mass to the point that the formation
of a neutron star is possible. The star also avoids the supergiant stage, during which
the core would lose angular momentum because of the core–envelope coupling.
Observationally, the presence of magnetars is often invoked in high-mass X-ray
binaries (e.g. [20]) and in particular for some persistent Be systems with longspin-period (1000 s) neutron stars. The issue is that, according to the standard
picture, after a short propeller phase, the neutron star enters the accretor stage
and its spin period quickly settles at an equilibrium value. To have an equilibrium
3 For the progenitor of SGR 1900+14, a lower mass of ∼17 M has been inferred, suggesting that
magnetars could form from stars with a wide spectrum of initial masses [32, 41].
4 They also found a candidate for the other member of the pre-supernova system: Wd1–5, a ∼9-M
runaway star that is escaping the cluster at high velocity and has a peculiar carbon excess that may
be due to the binary evolution.
P. Esposito et al.
3.2.6 Magnetars in Binary Systems
It is not clear whether magnetar exist in binary systems but, although it is possible
that magnetars to form have to sacrifice the companion, binarity may be an
important ingredient for the production of magnetars for at least two reasons.
Firstly, if magnetars descend from particularly massive stars (20 M ), the binary
interaction may prevent the formation of a black hole instead of a neutron star.
Secondly, in the case the magnetic fields of magnetars are formed through dynamo
amplification, even if magnetars form from stars with lower masses, a binary system
may help the stellar core to maintain the angular momentum necessary for the
dynamo mechanisms. On the other hand, in the fossil field scenario, in which
the neutron star magnetic field reflects that of the stellar precursor, progenitors
belonging to binary systems may be disfavoured, since magnetic hot stars in
compact binary systems are very rare [161].
Both scenarios for the origin of magnetar magnetic field require a very massive
progenitor, heavier than ∼20 M . For the fossil field, because there seems to be in
main sequence stars a trend of stronger magnetic fields with larger masses (Ferrario
and Wickramasinghe [70] and references therein). In the dynamo hypothesis,
because to produce neutron stars rotating fast enough to generate a magnetar field
via α-ω dynamo, a star more massive than ∼20–35 M star is necessary [98].
There are also observational facts that favour very massive stars as the magnetar
precursors. Considering the spatial distribution of the known magnetars, it has been
noted that their height on the Galactic plane is smaller than that of OB stars. This
suggests that they were produced by the most massive O stars [164]. Furthermore,
there are some tentative associations of magnetars with massive star clusters.
In the most convincing case, CXOU J164710.2–455216 in Westerlund 1, the
young age of the cluster (∼4 Myr) implies a progenitor with minimum mass of
∼40 M [158]. 3 To allow such a massive star to produce a neutron star, Clark et al.
[33] suggested a binary with (41 + 35) M stars and an orbital period shorter than
8 days. 4 The idea is that the binary interaction drives the primary in a Wolf–Rayet
star that through its powerful stellar winds loses mass to the point that the formation
of a neutron star is possible. The star also avoids the supergiant stage, during which
the core would lose angular momentum because of the core–envelope coupling.
Observationally, the presence of magnetars is often invoked in high-mass X-ray
binaries (e.g. [20]) and in particular for some persistent Be systems with longspin-period (1000 s) neutron stars. The issue is that, according to the standard
picture, after a short propeller phase, the neutron star enters the accretor stage
and its spin period quickly settles at an equilibrium value. To have an equilibrium
3 For the progenitor of SGR 1900+14, a lower mass of ∼17 M has been inferred, suggesting that
magnetars could form from stars with a wide spectrum of initial masses [32, 41].
4 They also found a candidate for the other member of the pre-supernova system: Wd1–5, a ∼9-M
runaway star that is escaping the cluster at high velocity and has a peculiar carbon excess that may
be due to the binary evolution.