Evidence for a Binary Companion to the Central Compact

Unique among neutron stars, 1E 1207.4−5209 is an X-ray pulsar with a spin period of 424 ms that contains at least two strong absorption features in its energy spectrum. This neutron star is positionally coincident with the supernova remnant PKS 1209−51/52 and has been identified as a member of the growing class of radioquiet compact central objects in supernova remnants. From previous observations with Chandra and XMMNewton, it has been found that the 1E 1207.4−5209 is not spinning down monotonically as is common for young, isolated pulsars. The spin frequency history requires either strong, frequent glitches, the presence of a fall-back disk, or a binary companion. Here, we report on a sequence of seven XMM-Newton observations of 1E 1207.4−5209 performed during a 40 day window between 2005 June 22 and July 31. Due to unanticipated variance in the phase measurements during the observation period that was beyond the statistical uncertainties, we could not identify a unique phase-coherent timing solution. The three most probable timing solutions give frequency time derivatives of +0.9, −2.6, and +1.6 × 10 Hz s (listed in descending order of significance). We conclude that the local frequency derivative during our XMM-Newton observing campaign differs from the longterm spin-down rate by more than an order of magnitude. This measurement effectively rules out glitch modP.M. Woods Dynetics, Inc. 1000 Explorer Blvd. Huntsville, AL 35806, USA Tel.: +1-256-964-4914 E-mail: [email protected] V.E. Zavlin Space Science Laboratory, NASA MSFC SD50, Huntsville, AL 35805, USA Tel.: +1-256-961-7463 E-mail: [email protected] G.G. Pavlov Pennsylvania State University, 525 Davey Lab, University Park, PA 16802, USA Tel.: +1-814-865-9482 E-mail: [email protected] els for 1E 1207.4−5209. If the long-term spin frequency variations are caused by timing noise, the strength of the timing noise in 1E 1207.4−5209 is much stronger than in other pulsars with similar period derivatives. Therefore, it is highly unlikely that the spin variations are caused by the same physical process that causes timing noise in other isolated pulsars. The most plausible scenario for the observed spin irregularities is the presence of a binary companion to 1E 1207.4−5209. We identified a family of orbital solutions that are consistent with our phase-connected timing solution, archival frequency measurements, and constraints on the companions mass imposed by deep IR and optical observations.