Follow-on Science Instrument

Contract NAS8-01129


Monthly Status Report No. 011

January 2003

HETG Science Theme: Isolated Neutron Stars

CREATOR: XV Version 3.10a  Rev: 12/29/94 (PNG patch 1.2) 
Quality = 90, Smoothing = 0CREATOR: XV Version 3.10a  Rev: 12/29/94 (PNG patch 1.2) 
Quality = 90, Smoothing = 0


Prepared in accordance with DR 972MA-002; DPD #972

Prepared for

National Aeronautics and Space Administration

Marshall Space Flight Center, Alabama 35812


Center for Space Research; Massachusetts Institute of Technology; Cambridge, MA 02139

Isolated Neutron Stars Research Progress


Introduction to Isolated Neutron Stars

         In the case of X-ray binaries we see a Neutron Star (NS) in a binary system with a companion and mass transferf rom the companion can lead to a bright source of X-rays. But what if the neutron star (presumably created in a core-collapse supernova event) does not have a companion? In this case we have an “isolated neutron star” (iNS). The iNS can be hot and radiate a smooth, featureless black body spectrum. The size, radius, of the neutron star defines the radiating surface area and radii are expected to be of order 10 km (for reference the Schwarzchild radius, R_Schw=2GM/c^2, for M=1.4 solar mass is ~4.2 km - this would appear to be a lower limit to actual NS radii.)

       Departures from a simple blackbody can arise if the neutron star has an atmosphere – due to the high gravity on the NS surface the height of this atmosphere is of order 0.01 to 1 cm – a very thin shell. The temperature of the surface, T*, the surface gravity, g, and the composition of the atmosphere (H, He, Fe, Si, etc.) then determine the spectrum and directionality of emitted radiation. Because of the high surface gravity, the emitted radiation, when seen by a distant observer, is red shifted with the apparent T_obs given by T_eff/(1+z); (1+z) is 1.31 for M=1.4, and R*=10.

If the neutron star has a substantial magnetic field and is spinning, it can interact with surrounding material to generate pulses of synchrotron radiation and be detected as a Pulsar from radio to X-ray.

[ Figure above is available in various formats: .png , .ps , .fig . -dd ]

Summary of Isolated Neutron Star GTO Observations and Activities

Three isolated Neutron Stars have been selected for inclusion in our HETG GTO observations to date and are listed in the Table below. Two other iNS are tabulated here as well: the Chandra-discovered point source in Cas A – which is also in our HETG GTO observation of Cas A (though that was not the main focus of the observation) and the LETG GTO (and later DDT) target RX J1856.5-3754 which obtained notoriety when the possibility of its being a “quark star” was suggested.

We have also worked on applying realistic atmosphere models to iNS specifically for the case of the Cas A point source in which context these models are discussed further below.







B field











4U 0142+61

8.7 s






Anomalous x-ray pulsar, BB+PL.

Young NS ? EW/E < 0.02





PSR B0656+14

384 ms


0.8 and


22 and



Two BB model.

Spin age 100 k yr.

H atmosphere model




PSR B0833-45,

Vela pulsar

89 ms






Spin age 11 k yr.

Magnetic H atmosphere model







Cas A X-ray Point Source


13 ms?


~ 10^10 ?







Inside Cas A remnant: BB w/edges


ATM H model fit, R_g.





RX J1856.5-3754

“A quark star?”







3.8 - 8.2



Drake et al., 2002.


Rot. blurred Si ATM model, R_g.



Neutron Star Atmosphere models and the Cas A Point Source

         Paper: M. Stage, “Recent results fitting ATM atmophere models to Chandra spectra of thermally radiating neutron stars” (COSPAR 2002)and thesis in preparation, 2003.

         The X-ray Point Source, XPS, in the Cas A supernova remnant (below left) was discovered by Chandra in its first scientific image as a small central point if emission in the remnant. It is likely a neutron star from the collapsed core of the progenitor star. Two absorption effects are seen and modelled in the ACIS spectrum of the XPS: the usual (but large) interstellar galactic absorption which cuts off low-energy photons (below right) and local absorption by Si and S in the remnant surrounding the XPS – this absorption is responsible for the extra notches created in the spectrum a little below and a little above 2 keV. With these absorption terms the spectrum can be fit with a simple blackbody of order T_obs = 5 MK and R_obs ~ 0.6 km. This radius is thought to be too small for NS models and so suggests a partial area of emission or NS atmosphere effects are involved.


CREATOR: XV Version 3.10a  Rev: 12/29/94 (PNG patch 1.2) 
Quality = 90, Smoothing = 0CREATOR: XV Version 3.10a  Rev: 12/29/94 (PNG patch 1.2) 
Quality = 90, Smoothing = 0

         The ATM atmosphere models have been used to create modelled surface spectra for NS for different atmosphere compositions, temperatures and surface gravity values. The Figure here shows three such models (solid black lines) corresponding, left-to-right, to i) a Si-ash model at T*=1.01 MK, ii) a Hydrogen-only model at T*=1.5 MK, and iii) an iron (Fe) atmosphere model at T* = 4.0 MK. Note the abundance of spectral features in the Si and Fe atmosphere models (the fluxes have been scaled and do not correspond to a specific physical system.) The effect of surface gravity, g, is also shown here: the solid-line models are for log(g) = 14.0 and for each a dotted-lined version is plotted corresponding to log(g) = 15.2; for reference log(g) is 14.4 for M=1.4, R*=10. The higher surface gravity moves the spectra to slightly higher energies (not unlike a small amount of pileup in ACIS!?…) With high precision data it maybe possible to constrain the g value and have a measure of the actual NS radius assuming a mass value.


The promise of observing such detailed structure as these Fe and Si models suggest has lead to observations of iNS with the HETG and LETG. As will be seen in the following, actual observations have yet to reveal any clear atmosphere signature in the spectra.



J. Madej, “Model atmospheres and X-ray spectra of bursting neutron stars”, ApJ, 376:161 (1991)

V.E. Zavlin et al., “Model neutron star atmospheres with low magnetic fields”, A&A, 315:141 (1996)

[ Figure created with the IDL routines: ins_plots.pro and ins_spectra.pro . -dd ]


B0656+14: Atmosphere on isolated Radio Pulsar?

Paper: H.L. Marshall and N.S. Schulz, “Using the High-resolution X-ray spectrum of PSR B0656+14 to constrain the chemical composition of the neutron star atmosphere”, ApJ, 574:377, 2002.

We observed PSR B0656+14 with GTO time using the LETG and HRC-S. The observed spectrum, below, is well modelled by the sum of two black bodies, the dominant one has T_obs = 0.8 MK and R_obs = 22 km. “No significant absorption features are found in the spectrum that might be expected from ionization edges of H or He or bound-bound transitions of Fe in magnetized atmopheres. … we conclude that the atmosphere is not dominated by Fe or other heavy elements that would be partially ionized at a temperature of 10^6 K.”


CREATOR: XV Version 3.10a  Rev: 12/29/94 (PNG patch 1.2) 
Quality = 90, Smoothing = 0




Vela Pulsar: Using everything Chandra’s got!

CREATOR: XV Version 3.10a  Rev: 12/29/94 (PNG patch 1.2) 
Quality = 90, Smoothing = 0Paper: G.G. Pavlov et al., “The X-ray spectrum of the Vela pulsar resolved with the Chandra X-ray Observatory”, ApJ, 552:L129, 2001. (These authors are at PSU and MPI Garching.)


The Vela pulsar was observed during Cycle 1 using three instrument combinations of Chandra: HETG/ACIS in CC mode (HETG GTO), ACIS-S no grating (ACIS GTO), and LETG/HRC-S (ACIS GTO). The image at right is from the ACIS-only observation and below it is the zeroth-order histogram from the HETG/ACIS CC observation. Note that although the pulsar itself is very bright, the surrounding pulsar wind nebula has a large total flux – this plus background events reduced the usefulness of the HETGS CC mode dispersed data.


The neutron star spectrum was modelled by a black body, T_obs = 1.49 MK and R_obs = 2.1 km, and a power-law relevant at high energies.


The featureless spectrum seen in the LETGS observation is consistant with a hydrogen NS atmosphere having an effective temperature at the surface of 0.9 MK and emission from the whole NS surface with R* ~ 10 to 13 km.



4U 0142+61: Anomalous X-ray Pulsar

Paper: A.M. Juett et al., “Chandra High-resolution spectrum of the anomanous X-ray pulsar 4U 0142+61”, ApJ, 568:L31, 2002.

CREATOR: XV Version 3.10a  Rev: 12/29/94 (PNG patch 1.2) 
Quality = 90, Smoothing = 0The anomalous X-ray pulsar 4U 0142+61 was observed with the HETGS in the Cycle 2 GTO porgram. The continuum spectrum, shown below along with residuals to the fit, is well fit by an absorbed power law plus blackbody with a photon index Tau = 3.3 and T = 4.8 MK.





“No evidence was found for emission or absorption lines. … If the source is a magnetar, then the absence of a proton cyclotron line strongly constrains magnetar atmosphere models and hence the magnetic field strength of the neutron star.” B fields in the 2 to 10 times 10^14 G range are excluded and this is consistant with a dipole field of B = 1.3 x 10^14 G at the polar cap estimated from pulsar spin-down.