2012ApJS..201...37K

Taking advantage of the high sensitivity of the Chandra X-ray Observatory's (CXO) Advanced CCD Imaging Spectrometer, we have conducted a snapshot survey of pulsars previously undetected in X-rays. We detected 12 pulsars and established deep flux limits for 11 pulsars. Using these new results, we revisit the relationship between the X-ray luminosity, L ^psr_X, and spin-down power, {dot}E. We find that the obtained limits further increase the extremely large spread in the non-thermal X-ray efficiencies, η^psr_X= L ^psr_X/{dot}E, with some of them being now below 10^–5. Such a spread cannot be explained by poorly known distances or by beaming of pulsar radiation. We also find evidence of a break in the dependence of L ^psr_X on {dot}E, such that pulsars become more X-ray efficient at {dot}E-10³⁵ erg/s. We examine the relationship between the γ-ray luminosity, L ^psr_γ, and {dot}E, which exhibits a smaller scatter compared to that in X-rays. This confirms that the very large spread in the X-ray efficiencies cannot be explained just by the beaming because the γ-ray emission is generally expected to be beamed stronger than the X-ray emission. Intriguingly, there is also an indication of a break in the L ^psr_γ{dot}E dependence at {dot}E erg/s, with lower-{dot}E pulsars becoming less γ-ray efficient. We also examine the distance-independent L ^psr_γ/L ^psr_X ratio as a function of {dot}E for a sample of γ-ray pulsars observed by CXO and find that it peaks at {dot}E erg/s, showing that the breaks cannot originate from poorly measured distances. We discuss the implications of our findings for existing models of magnetospheric emission and venues for further exploration.