Newborn magnetars as sources of gravitational radiation: constraints from high energy observations of magnetar candidates

Two classes of high-energy sources, the Soft Gamma Repeaters and the Anomalous X-ray Pulsars are believed to contain slowly spinning “magnetars,” i.e. neutron stars the emission of which derives from the release of energy from their extremely strong magnetic fields (>10¹⁵ G). The enormous energy liberated in the 2004 December 27 giant flare from SGR 1806-20 (～5×10⁴⁶ erg), together with the likely recurrence time of such events, points to an internal magnetic field strength of ≥10¹⁶ G. Such strong fields are expected to be generated by a coherent α?Ω dynamo in the early seconds after the Neutron Star (NS) formation, if its spin period is of a few milliseconds at most. A substantial deformation of the NS is caused by such fields and, provided the deformation axis is offset from the spin axis, a newborn millisecond-spinning magnetar would thus radiate for a few days a strong gravitational wave signal the frequency of which (～0.5–2 kHz range) decreases in time. This signal could be detected with Advanced LIGO-class detectors up to the distance of the Virgo cluster, where ≥1 yr^?1 magnetars are expected to form. Recent X-ray observations revealed that SNRs around magnetar candidates do not appear to have received a larger energy input than in standard SNRs (see Vink and Kuiper, Mon. Not. Roy. Astron. Soc. 319, L14 (2006)). This is at variance with what would be expected if the spin energy of the young, millisecond NS were radiated away as electromagnetic radiation and/or relativistic particle winds. In fact, such energy would be transferred quickly and efficiently to the expanding gas shell. This may thus suggest that magnetars did not form with the expected very fast initial spin. We show here that these findings can be reconciled with the idea of magnetars being formed with fast spins, if most of their initial spin energy is radiated through GWs. In particular, we find that this occurs for essentially the same parameter range that would make such objects detectable by Advanced LIGO-class detectors up to the Virgo Cluster. If our argument holds for at least a fraction of newly formed magnetars, then these objects constitute a promising new class of gravitational wave emitters.