A Multiwavelength Study of PSR B0628-28: The First Overluminous Rotation-powered Pulsar?

The ROSAT source RX J0630.8-2834 was suggested by positional coincidence to be the X-ray counterpart of the pulsar PSR B0628-28. This association, however, was regarded to be unlikely on the basis of the computed energetics of the putative X-ray counterpart. In this paper we report on multiwavelength observations of PSR B0628-28 made with the ESO NTT observatory in La Silla, the Lovell telescope at Jodrell Bank, and XMM-Newton. Although the optical observations do not detect any counterpart of RX J0630.8-2834 down to a limiting magnitude of V=26.1 mag and B=26.3 mag, XMM-Newton observations finally confirm it to be the pulsar's X-ray counterpart by detecting X-ray pulses with the radio pulsar's spin period. The X-ray pulse profile is not sinusoidal but characterized by a two-component pulse profile, consisting of a broad peak with a second narrow pulse leading the main pulse by ~144°. The fraction of pulsed photons is (39+/-6)% with no strong energy dependence in the XMM-Newton bandpass. The pulsar's X-ray spectrum is well described by a power law with photon index α=2.63^+0.23_-0.15. A composite Planckian-plus-power-law spectral model yields an interesting alternative that formally describes the observed energy spectrum equally well. Inferred from best fits are a blackbody temperature of ~1.7×10⁶ K and a projected blackbody radius of ~69⁺³⁰_-25 m, yielding a thermal flux contribution of ~20% within the 0.1-2.4 keV band. The pulsar's spin-down to X-ray energy conversion efficiency as obtained from the single-power-law spectral model is ~16%, assuming the distance inferred from the radio dispersion measure. If confirmed, PSR B0628-28 would be the first X-ray-overluminous rotation-powered pulsar identified among all ~1400 radio pulsars known today. The emission beam geometry of PSR B0628-28 is estimated from radio polarization data taken at 408 and 1400 MHz. A formal best fit of the 1400 MHz data yields α~11^deg for the inclination of the magnetic axis to the rotation axis and β~-3^deg for the impact angle of the line of sight. A combination of results obtained from 408 and 1400 MHz data, however, makes a nearly orthogonal solution, with α~70^deg and β~-12^deg the most likely one.