Search

The conversion of the Crab pulsar wind from one dominated by Poynting flux close to the star to one dominated by particle-borne energy at the termination shock is considered. The idea put forth by Coroniti (1990) and criticized by Lyubarsky & Kirk (2001) that reconnection in a striped wind is responsible, is generalized to include faster prescriptions for the a priori unknown dissipation rate. Strong acceleration of the wind is confirmed, and the higher dissipation rates imply complete conversion of Poynting flux into particle-borne flux within the unshocked wind.

We derive light curves for the hard gamma-ray emission, at energies up to several TeV, expected from the unique pulsar/Be-star binary system PSR B1259-63. This is the only known system in our galaxy in which a radio pulsar is orbiting a main sequence star. We show that inverse Compton emission from the electrons and positrons in the shocked pulsar wind, scattering target photons from the Be star, produces a flux of hard gamma-rays that should be above the sensitivity threshold of present day atmospheric Cerenkov detectors. Furthermore, we predict that the flux of hard gamma-rays produced via this mechanism has a characteristic variation with orbital phase that should be observable, and which is not expected from any other mechanism., To be published in Astroparticle Physics. 24 pages, 8 figures

The high-energy gamma-ray light curve of the binary system PSR B1259 -63, is computed using the approach that successfully predicted the spectrum at periastron. The simultaneous INTEGRAL and H.E.S.S. spectra taken 16 days after periastron currently permit both a model with dominant radiative losses, high pulsar wind Lorentz factor and modest efficiency as well as one with dominant adiabatic losses, a slower wind and higher efficiency. In this paper we shown how the long-term light curve may help to lift this degeneracy., Comment: 4 pages, to appear in proceedings of: Astrophysical Sources of High Energy Particles and Radiation, Torun (2005)

Flows in which energy is transported predominantly as Poynting flux are thought to occur in pulsars, gamma-ray bursts and relativistic jets from compact objects. The fluctuating component of the magnetic field in such a flow can in principle be dissipated by magnetic reconnection, and used to accelerate the flow. We investigate how rapidly this transition can take place, by implementing into a global MHD model, that uses a thermodynamic description of the plasma, explicit, physically motivated prescriptions for the dissipation rate: a lower limit on this rate is given by limiting the maximum drift speed of the current carriers to that of light, an upper limit follows from demanding that the dissipation zone expand only subsonically in the comoving frame and a further prescription is obtained by assuming that the expansion speed is limited by the growth rate of the relativistic tearing mode. In each case, solutions are presented which give the Lorentz factor of a spherical wind containing a transverse, oscillating magnetic field component as a function of radius. In the case of the Crab pulsar, we find that the Poynting flux can be dissipated before the wind reaches the inner edge of the Nebula if the pulsar emits electron positron pairs at a rate >1.E40 per second, thus providing a possible solution to the sigma-problem., Accepted for publication in ApJ

Motivated by recent unsuccessful efforts to detect the predicted flux of TeV gamma-rays from supernova remnants, we present a critical examination of the theory on which these predictions are based. Three crucial problems are identified: injection, maximum achievable particle energy and spectral index. In each case significant new advances in understanding have been achieved, which cast doubt on prevailing paradigms such as Bohm diffusion and single-fluid MHD. This indicates that more realistic analytical models, backed by more sophisticated numerical techniques should be employed to obtain reliable predictions. Preliminary work on incorporating the effects of anomalous transport suggest that the resulting spectrum should be significantly softer than that predicted by conventional theory., 8 pages, invited review presented at the 17th ECRS, Lodz, July 2000; to appear in Journal of Physics G: Nuclear and Particle Physics

We extend calculations of the shock acceleration mechanism applied to blazar jets. Electrons, assumed to undergo Bohm diffusion, are accelerated by the shock and radiate by synchrotron and inverse Compton scattering. We study the spectral variations in the X-ray and TeV regime as a function of flux and show that it can reproduce both the soft and the hard lags observed.

We investigate the effects of inverse Compton scattering by electrons and positrons in the unshocked winds of rotationally-powered binary pulsars. This process can scatter low energy target photons to produce gamma rays with energies from MeV to TeV. The binary radio pulsars PSR B1259-63 and PSR J0045-73 are both in close eccentric orbits around bright main sequence stars which provide a huge density of low energy target photons. The inverse Compton scattering process transfers momentum from the pulsar wind to the scattered photons, and therefore provides a drag which tends to decelerate the pulsar wind. We present detailed calculations of the dynamics of a pulsar wind which is undergoing inverse Compton scattering, showing that the deceleration of the wind of PSR B1259-63 due to `inverse Compton drag' is small, but that this process may confine the wind of PSR J0045-73 before it attains pressure balance with the outflow of its companion star. We calculate the spectra and light curves of the resulting inverse Compton emission from PSR B1259-63 and show that if the size of the pulsar wind nebula is comparable to the binary separation, then the gamma-ray emission from the unshocked wind may be detectable by atmospheric Cerenkov detectors or by the new generation of satellite-borne gamma-ray detectors such as INTEGRAL and GLAST. This mechanism may therefore provide a direct probe of the freely-expanding regions of pulsar winds, previously thought to be invisible., To be published in Astroparticle Physics. 27 pages, 5 figures

Observations of both gamma-ray burst sources and certain classes of active galaxy indicate the presence of relativistic shock waves and require the production of high energy particles to explain their emission. In this paper we review the basic theory of shock waves in relativistic hydrodynamics and magneto-hydrodynamics, emphasising the astrophysically interesting cases. This is followed by an overview of the theory of particle acceleration at such shocks. We summarise the applications to the astrophysics of relativistic jets and fireball models of gamma-ray-bursts., 43 pages, 5 figures, accepted for publication in Journal of Physics G: Nuclear and Particle Physics as a topical review

A homogeneous, synchro-self-Compton model (SSC) is used to model the spectra of gamma-ray emitting blazars. The constraints imposed by the rapid variability of the X-rays on a time scale of 1000 s can be met, but require a very large bulk Lorentz factors. These are not consistent with the bulk Lorentz factors implied by observations of apparent super-luminal motion in such sources. A model is also presented in which the target photons for the inverse Compton scattering stem from an external black body source. During a flare, this model predicts a lower relative amplitude of variation in the TeV regime as compared with the X-ray regime. It also gives rise to a steep spectrum in the TeV range. Finally, by varying the electron injection stochastically, the typical patterns of time variation of emission at different wavelengths are presented.

We present an acceleration mechanism based on multiple reflection of electrons by a shock front and apply it to explain the radio flare observed from SN1987A two days after explosion. The radiation mechanism is assumed to be synchrotron radiation from accelerated electrons which pass through the supernova shock front after having bounced off it many times during its early expansion into the progenitor’s stellar wind. Multiple reflection occurs because the wind contains a magnetic field wound into the form of a spiral by stellar rotation: the spiral becomes evermore tightly wound as the shock progresses outward and reflected electrons are unable to escape radially.For parameters appropriate to SN1987A, the shock becomes superluminal and reflections cease after a time of the order of days. The subsequent decay of the flare is a result of adiabatic losses in the expanding downstream plasma. The predicted spectrum is a power law I∝ν−1 above the synchrotron self‐absorption frequency, consistent with observations.

The theory of first order Fermi acceleration at relativistic shock fronts was developed initially in the test particle picture. However, the efficiency of the mechanism depends on the backreaction of the particles on the fluid. The investigation of stationary solutions which include this effect can in principle provide a means of estimating the efficiency. In the nonrelativistic case, stationary solutions turn out not to be relevant indicators of efficiency, basically because realistic acceleration timescales are too slow. Arguments are presented which indicate that this problem does not necessarily apply to shocks which move relativistically.

Following a technique developed by Saito, the positions and orientations of the polar rays visible in a photograph of the total solar eclipse of 12 November 1966 were examined to infer the length of an equivalent bar magnet representing the Sun's external magnetic field and the true distribution of polar rays from the two-dimensional projection seen in the photograph. In spite of the strong asymmetry in solar activity between the northern and southern solar hemispheres at the time of the eclipse, it was found that the magnetic field as delineated by the polar rays could be well represented by an extended dipole whose poles were each 0.43 R ⊙ from the center of the Sun and on the axis of rotation. The polar rays are distributed mainly in a circular zone 9° away from the pole and about 8° wide, with this zone bounded by empty regions at the pole and 15° to 20° away from it, beyond which there are again some rays, but not so many. These results are in good accord with expectation from Saito's analysis.

The collisional relaxation of an ion-beam in a plasma with an extremely strong magnetic field ( approximately 1013 G) is considered. The presence of a drift of the background electrons with respect to the background ions is included, giving rise to enhanced ion-sound fluctuations. The influence of the magnetic field on the orbit of the beam-ions enables these ions to scatter off the enhanced fluctuations. Once deflected from their motion parallel to the field the beam-ions undergo a relatively rapid collisional relaxation.

Magnetic fields of about 1012 G have now been measured in several astrophysical objects (Trümper etal. 1978, Wheaton et al. 1979, Mazets et al. 1981). In the case of pulsating X-ray sources, the presence of a companion star indicates that accretion is probably the energy source for the X-rays.

HARRISON1 has recently hypothesised that the Sun possesses a companion star, in order to explain an anomaly in the distribution on the sky of pulsars which lose speed very slowly. This is a controversial suggestion mainly because proper motion sky surveys down to at least 14th mag (ref. 2) should have detected such an object. However, one can imagine various objects which might fulfill the constraint of not being visible. Harrison suggested that crystallised white dwarfs, red dwarfs and black dwarfs could be in a bound orbit about the Sun, and neutron stars and black holes would more likely be in an unbound orbit, due to the supposedly explosive natures of their births. Whatever the luminosities of these objects3 they must produce an acceleration of the barycentre of the solar system to remove the pulsar anomaly. We discuss here the effect of this acceleration on the orbits of the ‘new’ comets, and how it rules out the possibility of a solar companion in a bound orbit.

Cyclotron lines have been observed in the X-ray spectra of two pulsed sources: Her X-1 (Trümper et al. 1978) and 4U 0115 + 63 (Wheaton et al. 1979). The generally accepted model for these objects involves an accretion flow from a companion star in a close binary system onto small regions close to the magnetic poles of a strongly magnetized neutron star. Immediately above the surface, matter is confined in an accretion column by the magnetic field.

The energy loss rate of fast protons in a strongly magnetized electron gas has been calculated using the Fokker‐Planck equation to describe the dominant effect of small‐angle Coulomb scattering. Stopping lengths or order 1 gm cm−2 are obtained for 30 MeV protons and the possibility of maser action at the gyrofrequency is indicated.

We derive expressions for the scattering and absorption cross sections of photons in a hot plasma including the effects of vacuum polarization in a strong magnetic field. Near the cyclotron resonance, these expressions depart significantly from previous cold plasma calculations. An approximate calculation of the radiative transfer is presented, and cyclotron absorption and emission-line fits are compared. This simplified transfer model suggests that an emission-line interpretation is slightly more probable at this stage.

Fokker-Planck coefficients are calculated to describe the interaction of a test particle distribution with an electron gas in a magnetic field of around 1012 G. For massive, classical test particles the Fokker-Planck equation is then solved numerically. Results are presented for the stopping length, the creation of an inverted distribution, and the final relaxation rate for a range of density and temperature values characteristic of the surface of an accreting neutron star. Differences of up to two orders of magnitude are found in comparison with other results in the literature. These arise because the gyro radius of the test particle is smaller than the Debye screening length of the electron gas.

Results are presented from detailed self-consistent models of accreting magnetized neutron star atmospheres, heated by the gradual deceleration of infalling protons via Coulomb encounters. The temperature and density gradients are calculated assuming momentum and energy balance, coupled with the radiative transfer for two polarizations. The cyclotron resonance effects were treated approximately. These models are characterized by power-law energy spectra, with single pulses at higher frequencies and multiple pulses at lower ones for some aspect angles, as well as a phase-dependent spectral index.

A new treatment of the calculation of the dielectric tensor of a magnetized plasma in an extremely strong field, such as occurs in magnetized neutron stars, is presented. An accurate numerical scheme for the evaluation of the hermitian and antihermitian parts of the tensor is used to calculate the refractive index and absorption cross section for magnetic fields and plasma temperatures typical of pulsating X-ray sources and y-ray burst sources.

Low-energy ions at interplanetary shocks have speeds similar to that of the background plasma. Under these conditions, the pitch-angle scattering which they are thought to undergo, and which is required if they are to be accelerated by the first-order Fermi mechanism, does not imply that the upstream distribution is nearly isotropic. The diffusion approximation must, therefore, be abandoned. A theory of anisotropic transport, previously used to treat acceleration at parallel relativistic shocks, is applied to this situation, leading to predictions of the pitch-angle distribution in the precursor to the shock. In situ measurements of the pitch-angle distribution of 30 keV ions, which have a speed of about six times that of the shock, show moderate anisotropies which agree well with the predictions of the diffusion approximation. In this case, the new theory demonstrates that the exact anisotropy is indeed well represented by the diffusion approximation. 34 references.

A Monte Carlo method is presented for the problem of acceleration of test particles at relativistic shocks. The particles are assumed to diffuse in pitch angle as a result of scattering off magnetic irregularities frozen into the fluid. Several tests are performed using the analytic results available for both relativistic and nonrelativistic shock speeds. The acceleration at relativistic shocks under the influence of radiation losses is investigated, including the effects of a momentum dependence in the diffusion coefficient. The results demonstrate the usefulness of the technique in those situations in which the diffusion approximation cannot be employed, such as when relativistic bulk motion is considered, when particles are permitted to escape at the boundaries, and when the effects of the finite length of the particle mean free path are important.