No Arabic abstract
We present a detailed investigation of atmospheres around accreting neutron stars with high magnetic field ($Bgtrsim 10^{12}$ G) and low luminosity ($Llesssim 10^{33}$ erg/s). We compute the atmospheric structure, intensity and emergent spectrum for a plane-parallel, pure hydrogen medium by solving the transfer equations for the normal modes coupled to the hydrostatic and energy balance equations. The hard tail found in previous investigations for accreting, non-magnetic neutron stars with comparable luminosity is suppressed and the X-ray spectrum, although still harder than a blackbody at the star effective temperature, is nearly planckian in shape. Spectra from accreting atmospheres, both with high and low fields, are found to exhibit a significant excess at optical wavelengths above the Rayleigh-Jeans tail of the X-ray continuum.
We construct models for strongly-magnetized neutron star atmospheres composed of mid-Z elements (carbon, oxygen and neon) with magnetic fields B=10^{12}-10^{13} G and effective temperatures Teff=(1-5)*10^6 K; this is done by first addressing the physics relevant to strongly-magnetized plasmas and calculating the equation of state and polarization-dependent opacities. We then obtain the atmosphere structure and spectrum by solving the radiative transfer equations in hydrostatic and radiative equilibrium. In contrast to hydrogen opacities at the relevant temperatures, mid-Z element opacities are dominated by numerous bound-bound and bound-free transitions. Consequently, temperature profiles are closer to grey profiles, and photosphere densities are lower than in the hydrogen case. Mid-Z element atmosphere spectra are significantly softer than hydrogen atmosphere spectra and show numerous absorption lines and edges. The atmosphere spectra depend strongly on surface composition and magnetic field but weakly on surface gravity. Absorption lines are primarily broadened by motional Stark effects and the (unknown) surface magnetic field distribution. Given the multiple absorption features observed from several isolated neutron stars, it is possible to determine, with existing X-ray data, the surface composition, magnetic field, temperature, and gravitational redshift; we present qualitative comparisons between our model spectra and the neutron stars 1E1207.4-5209 and RX J1605.3+3249. Future high-resolution X-ray missions such as Constellation-X will measure the gravitational redshift with high accuracy by resolving narrow absorption features, and when combined with radius measurements, it will be possible to uniquely determine the mass and radius of isolated neutron stars. (Abridged)
All the neutron star (NS) atmosphere models published so far have been calculated in the cold plasma approximation, which neglects the relativistic effects in the radiative processes, such as cyclotron emission/absorption at harmonics of cyclotron frequency. Here we present new NS atmosphere models which include such effects. We calculate a set of models for effective temperatures T_eff =1-3 MK and magnetic fields B sim 10^{10}-10^{11} G, typical for the so-called central compact objects (CCOs) in supernova remnants, for which the electron cyclotron energy E_{c,e} and its first harmonics are in the observable soft X-ray range. Although the relativistic parameters, such as kT_eff /(m_e c^2) and E_{c,e} /(m_e c^2), are very small for CCOs, the relativistic effects substantially change the emergent spectra at the cyclotron resonances, E approx sE_{c,e} (s=1, 2,...). Although the cyclotron absorption features can form in a cold plasma due to the quantum oscillations of the free-free opacity, the shape and depth of these features change substantially if the relativistic effects are included. In particular, the features acquire deep Doppler cores, in which the angular distribution of the emergent intensity is quite different from that in the cold plasma approximation. The relative contributions of the Doppler cores to the equivalent widths of the features grow with increasing the quantization parameter b_eff = E_{c,e}/kT_eff and harmonic number s. The total equivalent widths of the features can reach sim 150-250 eV; they increase with growing b_eff and are smaller for higher harmonics.
The crust of accreting neutron stars plays a central role in many different observational phenomena. In these stars, heavy elements produced by H-He burning in the rapid proton capture (rp-) process continually freeze to form new crust. In this paper, we explore the expected composition of the solid phase. We first demonstrate using molecular dynamics that two distinct types of chemical separation occur, depending on the composition of the rp-process ashes. We then calculate phase diagrams for three-component mixtures and use them to determine the allowed crust compositions. We show that, for the large range of atomic numbers produced in the rp-process ($Zsim 10$--$50$), the solid that forms has only a small number of available compositions. We conclude that accreting neutron star crusts should be polycrystalline, with domains of distinct composition. Our results motivate further work on the size of the compositional domains, and have implications for crust physics and accreting neutron star phenomenology.
Context: Many thermally emitting isolated neutron stars have magnetic fields larger than 10^13 G. A realistic cooling model that includes the presence of high magnetic fields should be reconsidered. Aims: We investigate the effects of anisotropic temperature distribution and Joule heating on the cooling of magnetized neutron stars. Methods: The 2D heat transfer equation with anisotropic thermal conductivity tensor and including all relevant neutrino emission processes is solved for realistic models of the neutron star interior and crust. Results: The presence of the magnetic field affects significantly the thermal surface distribution and the cooling history during both, the early neutrino cooling era and the late photon cooling era. Conclusions: There is a large effect of the Joule heating on the thermal evolution of strongly magnetized neutron stars. Both magnetic fields and Joule heating play a key role in keeping magnetars warm for a long time. Moreover, this effect is important for intermediate field neutron stars and should be considered in radio-quiet isolated neutron stars or high magnetic field radio-pulsars.
The flow of a matter, accreting onto a magnetized neutron star, is accompanied by an electric current. The closing of the electric current occurs in the crust of a neutron stars in the polar region across the magnetic field. But the conductivity of the crust along the magnetic field greatly exceeds the conductivity across the field, so the current penetrates deep into the crust down up to the super conducting core. The magnetic field, generated by the accretion current, increases greatly with the depth of penetration due to the Hall conductivity of the crust is also much larger than the transverse conductivity. As a result, the current begins to flow mainly in the toroidal direction, creating a strong longitudinal magnetic field, far exceeding an initial dipole field. This field exists only in the narrow polar tube of $r$ width, narrowing with the depth, i.e. with increasing of the crust density $rho$, $rpropto rho^{-1/4}$. Accordingly, the magnetic field $B$ in the tube increases with the depth, $Bpropto rho^{1/2}$, and reaches the value of about $10^{17}$ Gauss in the core. It destroys super conducting vortices in the core of a star in the narrow region of the size of the order of ten centimeters. Because of generated density gradient of vortices they constantly flow into this dead zone and the number of vortices decreases, the magnetic field of a star decreases as well. The attenuation of the magnetic field is exponential, $B=B_0(1+t/tau)^{-1}$. The characteristic time of decreasing of the magnetic field $tau$ is equal to $tausimeq 10^3$ years. Thus, the magnetic field of accreted neutron stars decreases to values of $10^8 - 10^9$ Gauss during $10^7-10^6$ years.