X-Ray and Gamma-Ray Observations of Pulsed Emission from Radio Pulsars.

This dissertation presents results from a search for pulsed emission from radio pulsars in the energy range 15 kev to 10 MeV using the UCSD/MIT Hard X-ray and Low Energy (gamma)-ray instrument on the HEAO-1 spacecraft. From a sample of twelve candidate radio pulsars, only the Crab pulsar was detected, and upper limits to the intensity of emission from the other candidates indicate that the majority (7 of 10) do not emit more than a few percent of their rotational energy loss in hard X-rays. The observations of the 18-200 keV pulsed emission from the Crab pulsar showed that the systematic variation of spectral slope as a function of pulse phase, discovered at lower energies (4-60 keV), continues up to (DBLTURN)200 keV, strengthening the case for the existence of at least two pulsed emission components: the emission from the two main pulse peaks and a separate interpulse emission connecting the two peaks. The interpulse emission contributes 22% of the total pulsed flux between 15 and 200 keV and has a harder spectrum than the spectrum common to the two pulse peaks. Combining these observations with others from infrared to (gamma)-ray energies leads to the following conclusions. (1) The phase-averaged pulsed spectrum, dominated by the emission from the two pulse peaks, requires at least three power law components with spectral breaks of (DBLTURN)0.5 in spectral index at (DBLTURN)1 keV and (DBLTURN)1MeV. (2) The interpulse emission has a spectrum better fit by either a thin, thermal bremsstrahlung model with kT = 151 (+OR -) 17 keV or a Comptonization model with kT = 26 (+OR-) 3 kev and scattering depth 5.2 (+OR-) 0.3 rather than the composite power law spectrum characteristic of the peaks. The thermal bremstrahlung model for the interpulse requires a large emission volume ((GREATERTHEQ) 10('24) cm('3)) and its validity can be tested with sensitive infrared, optical or (gamma)-ray observations of the spectrum, intensity and polarization of the emission. The Comptonization model requires either a large emission volume ((GREATERTHEQ) 10('24) cm('3)) or too high an input temperature ((GREATERTHEQ) 10('8)(DEGREES)K) and overestimates the observed optical intensity by a factor of (DBLTURN)10, but further theoretical study is required in order to incorporate the effects of the inferred, intense magnetic field ((DBLTURN)4 x 10('13) gauss). For the Vela pulsar, upper limits (1) fall below the low energy extrapolation of the > 50 MeV pulsed spectrum observed by the COS-B satellite, (2) are barely consistent with a single power law interpolation between the optical intensity and the intensity at 50 MeV, and (3) are consistent with predictions of a polar cap accleration model but not with a light cylinder model of synchrotron emission from a power law distribution of relativistic electrons. The reported 400-1100 keV pulsed emission from the radio pulsar PSR1822-09 is not confirmed. Hard X-ray (15-175 keV) upper limits for seven of the ten other pulsars, chosen for their proximity to Earth, their relatively high rates of rotational energy loss and their short periods, are only a few percent of their rotational energy losses, assuming a moment of inertia of 10('45) gm-cm('2) and a pulse duty cycle of (beta) = 0.4. The 15-175 keV Crab pulsed flux, excluding the interpulse, has a duty cycle of (DBLTURN)0.4 and has an intensity of 0.5% of the rate of rotational energy loss. Only Vela (0.006%) and PSR1929 + 10 (0.4%) have upper limits below the Crab value.