Have we detected continuum optical emission from any rotating neutron stars that do not have an accretion disk dominating the light? I ask because I know we have observed Doppler broadening of spectral lines from the rotation of even ordinary stars rotating at ordinary rates, but it seems to me that only the rapid rotation of a neutron star would be fast enough to make the Doppler broadening of the thermal continuum peak to be observable.

  • $\begingroup$ Per Wikipedia, a few have been observed by Hubble, but only close ones. They're very hard to observe. en.wikipedia.org/wiki/Neutron_star " Slow-rotating and non-accreting neutron stars are virtually undetectable; however, since the Hubble Space Telescope detection of RX J185635-3754, a few nearby neutron stars that appear only to emit thermal radiation have been detected." $\endgroup$
    – userLTK
    Dec 16, 2016 at 5:39
  • $\begingroup$ I am not sure what your issue with rotation is. The detectability of continuum radiation is unaffected by rotation. $\endgroup$
    – ProfRob
    Dec 16, 2016 at 9:28
  • $\begingroup$ Rotation is probably related to recent accretion, since accretion tends to spin up the neutron star, and the attendant heating from the accretion. $\endgroup$
    – Sean Lake
    Dec 16, 2016 at 20:56
  • $\begingroup$ Keep in mind that accretion disk and surface will emit at different bands, so there is no "dominion" of one over the other. It would also be nice, like @RobJeffries said, to specify what do you mean by continuum. There is NO astrophysical source which only emits lines (as far as a I know, at least). $\endgroup$
    – Py-ser
    Dec 17, 2016 at 8:50
  • 1
    $\begingroup$ What disk? Young neutron stars are not surrounded by disks. Disks occur when there is binary mass transfer onto old neutron stars. $\endgroup$
    – ProfRob
    Dec 17, 2016 at 21:20

1 Answer 1


Yes - quite a few isolated neutron stars have been observed, where any magnetospheric emission or accretion-related emission is either negligible or has been otherwise separated.

As you suspect, this emission is thermal in nature. Neutron stars are roughly approximated by black bodies but, like "normal" stars, they do have atmospheres and strong magnetic fields that modify the black body spectrum considerably. If the neutron star is isolated then the only reason that you might be able to detect thermal emission from the surface is that it is very young (less than about a million years after formation) and still hot. The difficulty is in ruling out any residual, pulsed magnetospheric component or a contribution from accretion of the interstellar medium itself.

A couple of examples to look at in the optical part of the spectrum would be Kulkarni & van Kerwijk (1998) who detected optical radiation from the counterpart to RX0720.4-3125; and Walter & Matthews (1997) who claimed a detection of thermal optical emission detection from the isolated neutron star RXJ185635-3754 (and see picture below).

There are several more examples of thermal emission detected at X-ray wavelengths from young, isolated neutron stars - the so-called X-ray dim neutron stars (see review by Mereghetti 2010).

Of course if you are extending your definition of continuum emission to include non-thermal (but not accretion-related) processes, then continuum emission is detected from all isolated pulsars. Synchrotron radiation is a continuum process and has been detected right across the electromagnetic spectrum in many pulsars.

A visible band HST image of the isolated neutron star RXJ185635-3754, attributable to Walter & Matthews 1997. HST image of RXJ185635-3754

  • $\begingroup$ It would be interesting to see how the magnetic fields would modify the BB-spectrum. I guess there is mainly theoretical work on this? $\endgroup$ Dec 16, 2016 at 12:29
  • $\begingroup$ Do any of the stars exhibit detectable Doppler broadening of the thermal peak? $\endgroup$
    – Sean Lake
    Dec 16, 2016 at 20:54
  • $\begingroup$ @SeanLake I misread your comment. What do you mean/envisage by "thermal peak".? Even spinning at 100 times a second, the rotational broadening of any spectral feature would amount to only about $\pm 2$%. This is not going to be detected in a slowly varying continuum. $\endgroup$
    – ProfRob
    Dec 17, 2016 at 14:56
  • $\begingroup$ That's exactly what I wanted to know, thanks. $\endgroup$
    – Sean Lake
    Dec 17, 2016 at 20:31

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .