A search for planetary companions around 800 pulsars from the Jodrell Bank pulsar timing programme

We have searched for planetary companions around 800 pulsars monitored at the Jodrell Bank Observatory, with both circular and eccentric orbits of periods between 20 d and 17 yr and inclination-dependent planetary masses from 10(-4) to 100 M-circle plus. Using a Bayesian framework, we simultaneously model pulsar timing parameters and a stationary noise process with a power-law power spectral density. We put limits on the projected masses of any planetary companions, which reach as low as 1/100th of the mass of the Moon (similar to 10(-4) M-circle plus). We find that two-thirds of our pulsars are highly unlikely to host any companions above 2 - 8 M-circle plus. Our results imply that fewer than 0.5 per cent of pulsars could host terrestrial planets as large as those known to orbit PSR B1257+12 (similar to 4 M-circle plus); however, the smaller planet in this system (similar to 0.02 M-circle plus) would be undetectable in 95 per cent of our sample, hidden by both instrumental and intrinsic noise processes, although it is not clear whether such tiny planets could exist in isolation. We detect significant periodicities in 15 pulsars; however, we find that intrinsic quasi-periodic magnetospheric effects can mimic the influence of a planet, and for the majority of these cases we believe this to be the origin of the detected periodicity. Notably, we find that the highly periodic oscillations in PSR B0144+59 are correlated with changes in the pulse profile and therefore can be attributed to magnetospheric effects. We believe the most plausible candidate for planetary companions in our sample is PSR J2007+3120.