Abstract. There has recently been growing evidence for the existence of neutron stars possessing magnetic fields with strengths that exceed the quantum critical field strength of 4.4 × 10 13 G, at which the cyclotron energy equals the electron rest mass. Such evidence has been provided by new discoveries of radio pulsars having very high spin-down rates and by observations of bursting gamma-ray sources termed magnetars. This article will discuss the exotic physics of this high-field regime, where a new array of processes becomes possible and even dominant, and where familiar processes acquire unusual properties. We review the physical processes that are important in neutron star interiors and magnetospheres, including the behavior of free particles, atoms, molecules, plasma and condensed matter in strong magnetic fields, photon propagation in magnetized plasmas, free-particle radiative processes, the physics of neutron star interiors, and field evolution and decay mechanisms. Application of such processes in astrophysical source models, including rotation-powered pulsars, soft gamma-ray repeaters, anomalous X-ray pulsars and accreting X-ray pulsars will also be discussed. Throughout this review, we will highlight the observational signatures of high magnetic field processes, as well as the theoretical issues that remain to be understood.

No abstract

No abstract

No abstract

Tidal interaction in a coalescing neutron star binary can resonantly excite the g-mode oscillations of the neutron star when the frequency of the tidal driving force equals the intrinsic g-mode frequencies. We study the g-mode oscillations of cold neutron stars using recent microscopic nuclear equations of state, where we determine self-consistently the sound speed and Brunt-Väisälä frequency in the nuclear liquid core. The properties of the g-modes associated with the stable stratification of the core depend sensitively on the pressure-density relation as well as the symmetry energy of the dense nuclear matter. The frequencies of the first ten g-modes lie approximately in the range of 10 − 100 Hz. Resonant excitations of these g-modes during the last few minutes of the binary coalescence result in energy transfer and angular momentum transfer from the binary orbit to the neutron star. The angular momentum transfer is possible because a dynamical tidal lag develops even in the absence of fluid viscosity. However, since the coupling between the g-mode and the tidal potential is rather weak, the amount of energy transfer during a resonance and the induced orbital phase error are very small.Resonant excitations of the g-modes play an important role in tidal heating of binary neutron stars.Without the resonances, viscous dissipation is effective only when the stars are close to contact. The resonant oscillations result in dissipation at much larger orbital separation. The actual amount of tidal heating depends on the viscosity of the neutron star. Using the microscopic viscosity, we find that the binary neutron stars are heated to a temperature ∼ 10 8 K before they come into contact.

We study implications for the apparent alignment of the spin axes, proper-motion directions, and polarization vectors of the Crab and Vela pulsars. The spin axes are deduced from recent Chandra X-ray Observatory images that reveal jets and nebular structure having definite symmetry axes. The alignments indicate these pulsars were born either in isolation or with negligible velocity contributions from binary motions. We examine the effects of rotation and the conditions under which spin-kick alignment is produced for various theoretical models of neutron star kicks. If the kick is generated when the neutron star first forms by asymmetric mass ejection or/and neutrino emission, then the alignment requires that the protoneutron star (and hence the precollapse stellar core) possesses an original spin with period P s much less than the kick timescale τ kick , thus spin-averaging the kick forces on the star. The kick timescale ranges from 100 ms to 10 seconds depending on whether the kick is hydrodynamically driven or neutrino-magnetic field driven. For hydrodynamical models, spin-kick alignment further requires the rotation period of an asymmetry pattern at the radius near shock breakout ( > ∼ 100 km) to be much less than τ kick < ∼ 100 ms; this is difficult to satisfy unless rotation plays a dynamically important role in the core collapse and explosion (corresponding to P s < ∼ 1 ms). Aligned kick and spin vectors are inherent to the slow process of asymmetric electromagnetic radiation from an off-centered magnetic dipole. We reassess the viability of this electromagnetic rocket effect, correcting a factor of 4 error in Harrison and Tademaru's calculation that increases the size of the effect. To produce a kick velocity of order a few hundred km s −1 requires that the neutron star be born with an initial spin close to one millisecond and that spindown due to r-mode driven gravitational radiation be inefficient compared to standard magnetic braking.

We carry out 2D viscous hydrodynamical simulations of circumbinary accretion using the moving-mesh code AREPO. We self-consistently compute the accretion flow over a wide range of spatial scales, from the circumbinary disk (CBD) far from the central binary, through accretion streamers, to the disks around individual binary components, resolving the flow down to 2% of the binary separation. We focus on equal-mass binaries with arbitrary eccentricities. We evolve the flow over long (viscous) timescales until a quasi-steady state is reached, in which the mass supply rate at large distancesṀ 0 (assumed constant) equals the time-averaged mass transfer rate across the disk and the total mass accretion rate onto the binary components. This quasi-steady state allows us to compute the secular angular momentum transfer rate onto the binary, J b , and the resulting orbital evolution. Through direct computation of the gravitational and accretional torques on the binary, we find that J b is consistently positive (i.e., the binary gains angular momentum), with l 0 ≡ J b /Ṁ 0 in the range of (0.4 − 0.8)a 2 b Ω b , depending on the binary eccentricity (where a b , Ω b are the binary semi-major axis and angular frequency); we also find that this J b is equal to the net angular momentum current across the CBD, indicating that global angular momentum balance is achieved in our simulations. In addition, we compute the time-averaged rate of change of the binary orbital energy for eccentric binaries, and thus obtain the secular rates ȧ b and ė b . In all cases, ȧ b is positive, i.e., the binary expands while accreting. We discuss the implications of our results for the merger of supermassive binary black holes and for the formation of close stellar binaries.

This note describes fitting formulae for the gravitational waveforms generated by a rapidly rotating neutron star (e.g., newly-formed in the core collapse of a supernova) as it evolves from an initial axisymmetric configuration toward a triaxial ellipsoid (Maclaurin spheroid ⇒ Dedekind ellipsoid). This evolution is driven by the gravitational radiation reaction (a special case of the CFS instability). The details and numerical results can found in [Lai & Shapiro, 1995, ApJ, 442, 259; Here referred as LS].I will use the units such that G = c = 1.The waveform (including the polarization) is given by Eq. (3.6) of LS. Since the waveform is quasi-periodic, I will give fitting formulae for the wave amplitude h (Eq. [3.7] of LS) and the quantity (dN/d ln f ) (Eq. [3.8] of LS; related to the frequency sweeping rate), from which the waveform h + (t) and h × (t) can be easily generated in a straightforward manner.Wave Amplitude: The waveform is parametrized by three numbers: f max is the maximum wave frequency in Hertz, M 1.4 = M/(1.4M ⊙ ) is the NS mass in units of 1.4M ⊙ , R 10 = R/(10 km) is the NS radius in units of 10 km. (Of course, the distance D enters the expression trivially.) It is convenient to express the dependence of h on t through f (the 1 This note was written in 1996. It was not intended for publication. Since I have been getting requests from people interested in GW data analysis about the waveform information, I thought it might be useful to put this note on gr-qc, so that I don't have to spend time looking for the TeX file every time I get a request.