News & Views | Published:


Jump-start for a neutron star

Radio emission from one of the neutron stars in the ‘double-pulsar’ system is strangely enhanced in two sections of its orbit — stimulated, perhaps, by radiation from its companion.

One of the most exciting discoveries in astronomy in recent times was that of the binary system1 J0737–3039, and its confirmation as a ‘double-pulsar’ system earlier this year2. Pulsars are rapidly spinning neutron stars that form during the supernova explosions of massive stars; although their masses tend to be slightly larger than that of our Sun, their radii are only about 15 km. For the first time, both neutron stars in this binary have been identified as radio pulsars — one that spins about its rotation axis every 22.7 milliseconds (which I shall refer to as ‘A’) and another (‘B’) that spins with a period of 2.77 seconds.

This duo promise to surpass even the original Nobel-prizewinning pulsar in a binary system3 as a testing ground for relativity, but they are also a fantastic laboratory for studying pulsar emission. The intense magnetic fields of pulsars accelerate charged particles around them, causing the emission of beams of radiation that sweep the sky like the rotating beams of a lighthouse. Already there are intriguing observations2 of the emission from the double-pulsar system — in particular that pulsar B seems to emit most strongly in two separate parts of its orbit. On page 919 of this issue, Jenet and Ransom4 offer an explanation for this strange effect, in a model that will have important implications for our understanding of this binary system.

The rotation periods of pulsars increase over time, reflecting the loss of rotational kinetic energy of the spinning neutron star as it emits a ‘wind’ of electromagnetic radiation along its emission beams. The difference in spin properties of the neutron stars in the double-pulsar binary means that their winds carry away energy at significantly different rates: the rate of loss of energy from A is some 3,000 times greater than that from B. This, and the compactness of the pulsars' orbit, implies that the energy carried in the respective winds from A and B is actually balanced inside the emission region of B (ref. 2). As a result, the energetics of A can be expected to dominate the system.

The panels of Fig. 1 show the geometry of the double-pulsar system, as seen from above the orbital plane. The two stars hurtle around their common centre of mass every 2.4 hours, at 0.1% of the speed of light. The two regions of strongest radio emission from pulsar B are indicated in orange in Fig. 1d. Because observers on the Earth are looking at the system nearly edge on, essentially in the same plane as the orbit, it is not surprising that the emission from B is strongest when it is closest to the Earth and A is furthest away. But it is not immediately obvious why there is a clear break in the emission between the two parts of the orbit.

Figure 1: The double-pulsar system J0737–3039.

The orbit of the pulsars, seen here from above the orbital plane, is so compact that it would fit inside the diameter of our own Sun (1.4 million kilometres). The radio emission from one of the pulsars, B, is known to be strongest in two particular regions of the orbit, and now Jenet and Ransom4 propose an explanation for why this is so. They assume that the other pulsar, A, emits radiation in a wide, hollow-cone beam. Panels a–d are snapshots of the pulsars' motion, showing the area swept out by A's beam. a, B intercepts A's beam and is stimulated to emit. b, This emission continues (orange band) until it enters the hollow midsection of A's beam and its emission is reduced. c, B is stimulated again as it enters the active part of A's beam for a second time. d, Once more, the emission is reduced when B moves out of A's beam. The orange bands representing stimulated emission from B match the regions of heightened emission seen in observations of the system. (Graphic derived from an animation at

Jenet and Ransom4 postulate that the emission from B is somehow stimulated — jump-started into action — when the lighthouse beam of A sweeps through B's emission region. The authors make the reasonable assumption that A's beam is a wide, hollow cone1 whose size and opening angle can be determined directly. It is then a relatively straightforward geometrical exercise to show that pulsar B intercepts A's beam at precisely the points of the orbit where increased emission is observed2. From current observations, the various angles in the system are constrained such that they fit two slightly different solutions of Jenet and Ransom's model, both of which produce the effect shown in Fig. 1.

As well as explaining observations, Jenet and Ransom's model makes testable predictions about the past and future visibility of the binary system. This is because the proposed geometry is strongly dependent on the relative orientation between A's emission beam and the line of sight from Earth. This angle varies with time through geodetic precession (a relativistic effect5 that occurs when the spin axis of an orbiting body is misaligned with the angular momentum axis of the binary system). The perturbing effect of B on the space-time of A causes the spin axis of A to precess around the angular-momentum axis. The strong gravitational field produced in the double-pulsar system means that A's spin axis precesses through a full 360° in 75 years. Similarly, B precesses every 71 years. These are the shortest geodetic precession periods ever observed and as a result the emission beams of A and B also move in and out of our line of sight within these timescales. This effect probably explains why the system was not visible during a previous survey of the sky over a decade ago6.

Using this precession rate in their two best-fit solutions, Jenet and Ransom predict that the emission beam of A will precess out of our line of sight in either 4.5 or 14 years, depending on the solution considered. Within the next year, as changes in A's beam geometry begin to accumulate, significant variations in the shape of that pulsar's radio pulses are expected; they should be sufficient to enable observers to decide between the two model solutions. As Jenet and Ransom point out, it is not yet certain whether the same precession effect will be observed for B because the wind from A might have caused its spin axis to align with the orbit.

Nature has provided a magnificent spectacle. Time, however, is most definitely of the essence as these two neutron stars may not be visible for much longer. Observational astronomers are now working feverishly to characterize this system further, taking data at many wavelengths across the electromagnetic spectrum. If we assume that the new model continues to describe the observations, the theoretical challenge is now to establish whether it is feasible to ‘jump-start’ a neutron star and what physical processes could cause this to occur.


  1. 1

    Burgay, M. et al. Nature 426, 531–533 (2003).

  2. 2

    Lyne, A. G. et al. Science 303, 1153–1157 (2004).

  3. 3

    Taylor, J. H. Rev. Mod. Phys. 66, 711–719 (1994).

  4. 4

    Jenet, F. A. & Ransom, S. M. Nature 428, 919–921 (2004).

  5. 5

    Barker, B. M. & O'Connell, R. F. Astrophys. J. 199, L25–L26 (1975).

  6. 6

    Lyne, A. G. et al. Mon. Not. R. Astron. Soc. 295, 743–755 (1998).

Download references

Author information

Rights and permissions

Reprints and Permissions

About this article

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.