ON THE DETECTABILITY OF PULSARS IN CLOSE BINARY-SYSTEMS

Globular clusters appear to contain a large number of radio pulsars and low mass X-ray binaries (LMXBs). These are believed to have been formed by tidal interaction of degenerate stars with the field stars. Some analyses of the cluster pulsar surveys indicate there is a large population of pulsars in clusters, with the discovered pulsars forming only the tip of the iceberg. However, most of the current pulsar searches are severely biased against finding pulsars in tight binaries because of the changing velocity of the pulsar. Recently several groups have attempted to improve this situation by using pulsar analysis codes which assume that the pulsar frequency changes linearly with time. This corresponds to assuming the acceleration of the pulsar is constant over the observation interval. A related problem in the search for coherent pulsations from LMXBs. Detection of such pulsations would provide the definitive proof that the LMXBs are the progenitors of the cluster pulsars. In this paper, we address both these issues. We introduce an efficiency factor-gamma to describe the amount by which the Fourier amplitude of the pulsar signal is reduced by binary motion and explore the extent to which this amplitude can be recovered by using acceleration codes. We present the results of numerical calculations of gamma as a function of pulsar spin period, binary orbital period, companion mass, and orbital inclination. We examine the advantage of using acceleration codes over standard pulse search techniques in this parameter space. We suggest strategies for maximizing the probability of finding pulsars in tight binaries. Finally, we show how current surveys are already placing interesting constraints on the tidal capture model of cluster pulsar formation.