CSR Home Page:
MIT Chandra Science Center
Chandra Science:

Norberts's Isolated Neutron Stars Page

The MIT/Chandra Science Neutron Star Team

Prof. Claude Canizares
Herman L. Marshall
Norbert S. Schulz


Neutron stars are extremely dense compact objects with masses of about 1.4 Msun (although masses were proposed to go up to 3) and a star radius of 10 to 20 km, depending on the equations of state proposed for the interior. Internal densities vary between 10^6 and 5 times 10^11 g/cm3 in the outer crust, a thin (300 to 800 m) solid region in which a Coulomb lattice of heavy nuclei coexists with a relativistic electron gas and 10^15 g/cm3 in the core containing mainly superfluid neutrons coexisting with electrons and protons. Above the outer crust there is a surface region with densities below 10^6 g/cm3 (Shapiro, S.L., and Teukolsky, S.A. 1983 "Black Holes, White Dwarfs, and Neutron Stars" (NY: John Wiley&Sons) ).

Neutron Stars are formed during supernova explosions and represent massive stars in their final stage of evolution. First evidence for the formation of neutron stars in supernova explosions came from the discoveries of the pulsars in the Crab and Vela supernova remnants. Born with gigantic magnetic fields (< 10^15 G) and rotational periods up to just a few milliseconds they loose rotational energy through radiation presumably in the radio domain. In the case of the Crab pulsar the implied energy loss through radio emission is roughly the same as the energy required to power the Crab nebula. Today we know of about 700 rotaionally powered radio-pulsars, however from which only 13 are associated with supernova remnants.

Pulsed radio emission is not the only means of energy dissipation in rotation powered pulsars (W. Becker, Ph.D. thesis, MPE Report 260, 1995) ). X-ray emission from 27 pulsars have been detected by the ROSAT and ASCA observatories, and several are also known to also radiate in the optical and gamma domain. The Geminga pulsar, for example is well established as an X-ray and gamma-ray source, but not as a source for radio emission. The soft X-ray luminosity for all the X-ray pulsars observed with ROSAT seem to have a linear relationship with the spin-down luminosity ( Becker et al. 1997 ). This now established relationship suggests that the bulk of the observed pulse X-ray emission is indeed emitted at the expense of rotational energy of the neutron star and thus of non-thermal origin. However, in general the detailed physics of energy dissipation through pulsar magnetospheres, especially if it comes to broad band synchrotron emission from the radio to the gamma-ray band, is yet poorly understood. Detailed highly resolved (spatial and spectral) spectroscopic X-ray studies will therefore greatly contribute to our understanding of pulsar magnetospheric emission processes.

X-rays from isolated (non-accreting) neutron stars are not entirely pulsed, and sometimes no pulsed fraction has been detected at all, i.e there is a significant fraction in the X-ray luminosity, which is not of magnetospheric origin. The spectrum of this radiation is a blackbody, that is of thermal nature. During the stellar collapse phase at the supernova explosion the temperature of the forming compact remnant reaches several 100 Billion degrees, which then rapidly (within hours od a few days) cooles down by a factor of more than 100 through neutrino emission. Depending on neutron star cooling models the neutron star surface will stay above temperatures of a few million degrees for about 10^5 years in which it will cool down by mostly emitting X-rays (during the first 10^4 years cooling may be still dominated by neutrino emission). A recent result obtained with ROSAT show that neutron star surface temperatures derived from soft X-ray emission are quite compatible with the predictions of standard cooling models. However, this result was derived by neglecting the effect of radiation transport through the neutron star atmosphere, which is the dense and anisotropic layer above the neutron star crust (see above). Because the radius of the neutron star is only 3 to 4 Schwarzschild-radii relativistic effects, like the gravitational redshift, will also alter the emitted spectrum.

The influence of this layer on the thermal emsission is the main focus of the investigation. Gravity at the neutron star surface the is very high (g = 10^14 cm/cm2) and therefore we expect a scale height of an atmospheric layer of only a few centimeters to maybe meters (this is the length, where the atmospheric pressure drops to 1/e). At high photon energies electron scattering will be incoherent and not isotropic and Compton scattering will influence radiative transport. The ion content and structure (remember that all ions will be extremely magnetized) is currently also very uncertain. However, new improved model atmospheres ( Rajagopal et al. 1997 , Zavlin et al. 1996 , Pavlov et al. 1995 ) predict a variety of signatures in the emission spectrum detectable with the HETGS and LETGS. Here it is crucial to accurately estimate the abundance of heavy elements, like Fe, in the presumably hydrogen dominated atmosphere. Highly magnetized Fe is expected to imprint narrow Landau-absoprtion features in the emergent soft X-ray spectrum. Observations of isolated neutron stars then involves the following key issues:

Perform high resolution spectroscopy of pulsar magnetospheric emission
Detect narrow features in soft thermal emission
Identify features based on existing model atmospheres
Refine existing neutron star model atmospheres
Resolve emission from pulsar and possible synchtoron nebula

Pulsars in Synchrotron Nebulae

TITLE: An HETGS observation of the Vela Pulsar (B0833-43)

The Vela pulsar and its remnant belongs to the closest ones known with a projected distance of 500 pc. Although its X-ray luminosity is more than a factor 10^4 lower than the one observed from the crab pulsar, it still offers enough outflow to power a synchrotron nebula. In fact, from all other synchrotron nebula candidates, Vela is the faintest and thus only visible because of its closeness. Einstein and Rosat observations showed, that although wisps as observed in the Crab are present, there are no shocks in the vicinity indicating a much gentler outflow. In this respect the Vela system entirely differs from the crab nebula and thus is an essential link to test charge injection dynamics. The pulsar itself has a period of 89.29 ms and a spin down age of roughly 11000 years, which is still well within the expected time scale to expect thermal readiation form a 10^6 K hot neutron star surface. The X-ray spectrum is dominanted by a steep power law, with a soft unpulsed component.

Isolated Neutron Stars

TITLE: High Resolution Soft X-ray Spectroscopy of B0656+14

PSR 0656+14 may be one of the closest cooling neutron stars to earth with a distance between 250 and 500 pc. Suggesting a distance of 250 pc, i.e half the distance to the Vela pulsar, it still emits a blackbody spectrum of 9 times10^5 K at an X-ray luminosity of 6 times 10^{32} erg/s. With a pulse period of 384.87 s and a spin down age of about 10$^5$ years it is slow and quite old. The moderate temperature, a still quite hight surface magnetic field strength (log B = 12.7 Gauss) and its proximity make it a prime candidate for soft X-ray spectroscopy. Its high energy spectrum is basically unknown.


Rajagopal et al. 1997
Zavlin et al. 1996
Becker et al. 1997
W. Becker, Ph.D. thesis, MPE Report 260, 1995
Pavlov et al. 1995

Norbert's Chandra Science page