Parkes radio searches of Fermi gamma-ray sources and millisecond pulsar discoveries

In a search with the Parkes radio telescope of 56 unidentified Fermi-LAT gamma-ray sources, we have detected 11 millisecond pulsars (MSPs), 10 of them discoveries, of which five were reported in Kerr et al. (2012). We did not detect radio pulsations from another six pulsars now known in these sources. We describe the completed survey, which included multiple observations of many targets done to minimize the impact of interstellar scintillation, acceleration effects in binary systems, and eclipses. We consider that 23 of the 39 remaining sources may still be viable pulsar candidates. We present timing solutions and polarimetry for five of the MSPs, and gamma-ray pulsations for PSR J1903-7051 (pulsations for five others were reported in the second Fermi-LAT catalog of gamma-ray pulsars). Two of the new MSPs are isolated and five are in >1 d circular orbits with 0.2-0.3 Msun presumed white dwarf companions. PSR J0955-6150, in a 24 d orbit with a ~0.25 Msun companion but eccentricity of 0.11, belongs to a recently identified class of eccentric MSPs. PSR J1036-8317 is in an 8 hr binary with a >0.14 Msun companion that is probably a white dwarf. PSR J1946-5403 is in a 3 hr orbit with a >0.02 Msun companion with no evidence of radio eclipses.

PSR J2030+3641: radio discovery and gamma-ray study of a middle-aged pulsar in the now identified Fermi-LAT source 1FGL J2030.0+3641

In a radio search with the Green Bank Telescope of three unidentified low Galactic latitude Fermi-LAT sources, we have discovered the middle-aged pulsar J2030+3641, associated with 1FGL J2030.0+3641 (2FGL J2030.0+3640). Following the detection of gamma-ray pulsations using a radio ephemeris, we have obtained a phase-coherent timing solution based on gamma-ray and radio pulse arrival times that spans the entire Fermi mission. With a rotation period of 0.2 s, spin-down luminosity of 3e34 erg/s, and characteristic age of 0.5 Myr, PSR J2030+3641 is a middle-aged neutron star with spin parameters similar to those of the exceedingly gamma-ray-bright and radio-undetected Geminga. Its gamma-ray flux is 1% that of Geminga, primarily because of its much larger distance, as suggested by the large integrated column density of free electrons, DM=246 pc/cc. We fit the gamma-ray light curve, along with limited radio polarimetric constraints, to four geometrical models of magnetospheric emission, and while none of the fits have high significance some are encouraging and suggest that further refinements of these models may be worthwhile. We argue that not many more non-millisecond radio pulsars may be detected along the Galactic plane that are responsible for LAT sources, but that modified methods to search for gamma-ray pulsations should be productive -- PSR J2030+3641 would have been found blindly in gamma rays if only >0.8 GeV photons had been considered, owing to its relatively flat spectrum and location in a region of high soft background.

Three Millisecond Pulsars in FERMI LAT Unassociated Bright Sources

We searched for radio pulsars in 25 of the non-variable, unassociated sources in the Fermi LAT Bright Source List with the Green Bank Telescope at 820 MHz. We report the discovery of three radio and gamma-ray millisecond pulsars (MSPs) from a high Galactic latitude subset of these sources. All of the pulsars are in binary systems, which would have made them virtually impossible to detect in blind gamma-ray pulsation searches. They seem to be relatively normal, nearby (<=2 kpc) millisecond pulsars. These observations, in combination with the Fermi detection of gamma-rays from other known radio MSPs, imply that most, if not all, radio MSPs are efficient gamma-ray producers. The gamma-ray spectra of the pulsars are power-law in nature with exponential cutoffs at a few GeV, as has been found with most other pulsars. The MSPs have all been detected as X-ray point sources. Their soft X-ray luminosities of ~10^{30-31} erg/s are typical of the rare radio MSPs seen in X-rays.

Out of the frying pan: a young pulsar with a long radio trail emerging from SNR G315.9-0.0

The faint radio supernova remnant SNR G315.9-0.0 is notable for a long and thin trail that extends outward perpendicular from the edge of its approximately circular shell. In a search with the Parkes telescope we have found a young and energetic pulsar that is located at the tip of this collimated linear structure. PSR J1437-5959 has period P = 61 ms, characteristic age tau_c = 114 kyr, and spin-down luminosity dE/dt = 1.4e36 erg/s. It is very faint, with a flux density at 1.4 GHz of about 75 uJy. From its dispersion measure of 549 pc/cc, we infer d ~ 8 kpc. At this distance and for an age comparable to tau_c, the implied pulsar velocity in the plane of the sky is V_t = 300 km/s for a birth at the center of the SNR, although it is possible that the SNR/pulsar system is younger than tau_c and that V_t > 300 km/s. The highly collimated linear feature is evidently the pulsar wind trail left from the supersonic passage of PSR J1437-5959 through the interstellar medium surrounding SNR G315.9-0.0.

Pulsar Timing for the Fermi Gamma-ray Space Telescope

We describe a comprehensive pulsar monitoring campaign for the Large Area Telescope (LAT) on the {em Fermi Gamma-ray Space Telescope} (formerly GLAST). The detection and study of pulsars in gamma rays give insights into the populations of neutron stars and supernova rates in the Galaxy, into particle acceleration mechanisms in neutron star magnetospheres, and into the engines driving pulsar wind nebulae. LATs unprecedented sensitivity between 20 MeV and 300 GeV together with its 2.4 sr field-of-view makes detection of many gamma-ray pulsars likely, justifying the monitoring of over two hundred pulsars with large spin-down powers. To search for gamma-ray pulsations from most of these pulsars requires a set of phase-connected timing solutions spanning a year or more to properly align the sparse photon arrival times. We describe the choice of pulsars and the instruments involved in the campaign. Attention is paid to verifications of the LAT pulsar software, using for example giant radio pulses from the Crab and from PSR B1937+21 recorded at Nanc{c}ay, and using X-ray data on PSR J0218+4232 from XMM-Newton. We demonstrate accuracy of the pulsar phase calculations at the microsecond level. Data Table 1 is only available in electronic form at the CDS via anonymous ftp to cdsarc.u-strasbg.fr (130.79.128.5) or via http://cdsweb.u-strasbg.fr/cgi-bin/qcat?J/A+A/ .

The magnetar 1E 1547.0-5408: radio spectrum, polarimetry, and timing

We have investigated the radio emission from the anomalous X-ray pulsar 1E 1547.0-5408 (PSR J1550-5418) using the Parkes telescope and the Australia Telescope Compact Array. The flux density of the pulsar is roughly the same between 1.4 and 45 GHz, but shows time variability. The radiation is nearly 100% linearly polarized between frequencies of 45 and 3.2 GHz, but from 2.3 to 1.4 GHz it becomes increasingly more depolarized. The rotation measure of -1860 rad/m^2 is the largest for any known pulsar, and implies an average magnetic field strength along the line of sight of 2.7 microG. The pulse profiles are circularly polarized at all frequencies observed, more so at lower frequencies, at the ~15% level. The observed swing of the position angle of linear polarization as a function of pulse phase suggests that in this neutron star the rotation and magnetic axes are nearly aligned, and that its radio emission is only detectable within a small solid angle. Timing measurements indicate that the period derivative of this 2 s pulsar has increased by nearly 40% in a 6-month period. The flat spectrum and variability in flux density and pulse profiles are reminiscent of the properties of XTE J1810-197, the only other known radio-emitting magnetar, and are anomalous by comparison with those of ordinary radio pulsars.

The variable radio-to-X-ray spectrum of the magnetar XTE J1810-197

We have observed the 5.54s anomalous X-ray pulsar XTE J1810-197 at radio, millimeter, and infrared (IR) wavelengths, with the aim of learning about its broad-band spectrum. At the IRAM 30m telescope, we have detected the magnetar at 88 and 144GHz, the highest radio-frequency emission ever seen from a pulsar. At 88GHz we detected numerous individual pulses, with typical widths ~2ms and peak flux densities up to 45Jy. Together with nearly contemporaneous observations with the Parkes, Nancay, and Green Bank telescopes, we find that in late 2006 July the spectral index of the pulsar was -0.5<alpha<0 over the range 1.4-144GHz. Nine dual-frequency Very Large Array and Australia Telescope Compact Array observations in 2006 May-September are consistent with this finding, while showing variability of alpha with time. We infer from the IRAM observations that XTE J1810-197 remains highly linearly polarized at millimeter wavelengths. Also, toward this pulsar, the transition frequency between strong and weak scattering in the interstellar medium may be near 50GHz. At Gemini, we detected the pulsar at 2.2um in 2006 September, at the faintest level yet observed, K_s=21.89+-0.15. We have also analyzed four archival IR Very Large Telescope observations (two unpublished), finding that the brightness fluctuated within a factor of 2-3 over a span of 3 years, unlike the monotonic decay of the X-ray flux. Thus, there is no correlation between IR and X-ray flux, and it remains uncertain whether there is any correlation between IR and radio flux.