Gold T. Rotating Neutron Stars As the Origin of the Pulsating Radio Sources

Home , Thomas Gold

CC/NUMBER 8 ® FEBRUARY 22, 1993 This Week's Citation Classic Gold T. Rotating neutron stars as the origin of the pulsating radio sources. Nature 218:731-2, 1968; and Gold T. Rotating neutron stars and the nature of pulsars. Nature 221:25-7, 1969. [Center for Radiophysics and Space Research: and Cornell-Sydney University Astronomy Center. Cornell University, Ithaca. NY] Enigmatic, rapidly pulsating radio sources were preferred to attribute these sources to interpreted as rotating beacons sweeping over distant galaxies, not to stellar objects. the Earth, radiating out from rapidly rotating, The pulsars now seemed to repre- extremely collapsed stars ("neutron stars"). Nu- merous detailed predictions followed from this sent just the stellar objects I had dis- model, and all were soon verified. The model cussed then. Calculations existed for also accounted closely for the previously unex- the collapsed "neutron stars" that in- plained luminosity of the Crab Nebula. [The dicated approximately their size, as SCI® indicates that these papers have been small as a few kilometers, and their cited in more than 240 and 150 publications, mass, on the order of a solar mass.4 respectively.] Astronomers generally thought that even if they existed, they could never be discovered. However hot, a star so small could not be seen at astronomi- The Nature of Pulsars cal distances. But they had not con- sidered the energy concentration re- Thomas Gold sulting from the collapse: enormous Center for Radiophysics magnetic field strengths and a spin and Space Research energy quite comparable with the en- Cornell University tire content of nuclear energy of the Ithaca, NY 14853 star before its collapse. With these considerations it was not unreason- The Cambridge radio astronomy able to expect them to be observable. group announced, in two papers in I had another clue to the nature of February and April 1968, the startling the new objects: While there was some observations of rapidly recurring short irregularity in the pulse-to-pulse tim- radio pulses from four sources in the ing, the long-term accuracy of these sky; each had a constant pulse repeti- "clocks" was enormously better than tion frequency ranging from 1/4 sec- a statistical addition of the pulse ir- ond for the fastest to 1.33 seconds for regularities would have allowed. the slowest.1'2 I proposed the model of a rapidly How could any object at astronomi- spinning neutron star, which, as a re- cal distances transmit sufficient en- sult of some asymmetries, sent out a ergy in brief pulses, so as to be ob- strong beam of radiation from one re- servable here? The source had to be gion of longitude. This beam would small, so that the light time across it sweep over the Earth once in each would not lengthen the pulses beyond rotation period and would be seen as the few milliseconds observed. A cal- a short pulse. The underlying long- culation gave almost unbelievably high term accuracy would then be produced values of the required intensity at the by the spin of the object, while short- source. term timing fluctuations would just At a conference in London in 19513 I reflect details of the emitting mecha- had argued that dense, collapsed stars nism. I compared it to the rotating would be ideally suited to emit strong beacon of a lighthouse, whose lamp, radio signals, since their magnetic hanging from the rotating shaft, could fields may be enormously strength- wobble a little; the long-term accuracy ened by the collapse and extend out would still be that of the rotation of the into low density space. But such ob- shaft. jects would be expected to show rapid My submission to Nature explained time variations, not present for the these points and predicted that pul- radio sources then known. I therefore sars might be found in the location of

8 CURRENT CONTENTS® ®1993 by lSI® supernova explosions, since neutron to date, a 33 millisecond repetition stars may be products of that process; frequency, was found in the Crab Nebula, the site of the most recent I also noted that fresh neutron stars 7 could be expected to rotate consider- supernova known (914 years old). ably faster than the pulsars so far dis- Within a day the expected slowdown covered, and one should therefore was also observed. search for shorter periods. Moreover, The Crab Nebula luminosity of about if rotation was the energy source, one 100,000 times that of the sun now had should be able to observe a slight an explanation. The observed value of slowdown in this long-term clock. the slowdown and the moment of iner- A conference on pulsars was orga- tia of a neutron star implied an energy nized in New York for May 20 and 21. loss that closely matched the observed Many speakers and some of the origi- luminosity of the nebula. Evidently, nal discovery group were invited. I energetic particles, fed by the rota- had sent my paper to the organizers tional energy, were responsible. Atthis with the request for a five minute slot, stage there were no doubts left about but this was turned down: "The sug- the explanation, and I submitted this gestion was so outlandish that if this result in the second paper to Nature, was admitted there would be no end to received there on December 10. the number of other suggestions that A great new era in astronomy had would equally have to be allowed."5 begun. A world of very high density, Meanwhile, the editor of Nature ob- very high energy concentration had viously thought otherwise. The paper, opened up. Here relativity theory was received on May 20, appeared on May supreme, not a minor correction to 25, probably a record in publication Newtonian gravitation. It was a world speed for the journal. not only of strong radio pulses, but of The New York conference was a fi- X-rays and high energy particles. The asco, dominated by a report of star- knowledge that neutron stars with tling new observations that turned out masses and densities close to the rela- to be entirely false and by theoretical tivistic configuration of a black hole models of radially pulsating white really existed meant that black holes dwarf stars. probably also existed. Pulsar observa- But I did not have to wait long for the tions have given us by far the most confirmations of my theory. By Octo- accurate confirmations of general rela- ber of that year, the Australians re- tivity, and the first observations of the process of emission of gravitational ported the detection of a pulsar with 3 an 89 millisecond period, the shortest waves, as well as many other new as- then known, and that in a location of a tronomical data. rather recent (20,000 year old) super- In the citation for the award to me, in nova explosion.6 In November the 1985, of the Gold Medal of the Royal Arecibo Observatory in Puerto Rico, a Astronomical Society of Great Britain, facility under my direction, made the my identification of the nature of pul- clinching discovery: the fastest pulsar sars was the lead item.

1. Hewish A, Bell S J. Pilkington J D H, Scott P F & Collins R A. Observation of a rapidly pulsating radio source. Nature 217:709-13, 1968. (Cited 250 times.) 2. Pilkington J D H, Hewish A, Bell S J & Cole T W. Observations of some further pulsed radio sources. Nature 218:126-9. 1968. 3. Gold T. The origin of cosmic radio noise. (Lang K R & Gingerich O, eds.) A source book in astronomy and astrophysics. 1900-1975. Cambridge, MA: Harvard University Press, 1979. p. 782-5. 4. Oppenheimer J R & Volkoff G M. On massive neutron cores. Phys. Rev. 55:374-81, 1939. (Cited 310 times since 1945.) 5. Cameron A G W. Personal communication, 1968. 6. Large M I, Vaughan A E & Mills B Y. A pulsar supernova association? Nature 220:340-1. 1968. 7. Cornelia J M, Craft H D, Lovelace R V E, Sutton J M & Leonard Tyler G. Crab Nebula pulsar NP 0532. Nature 221:453-4, 1969. 8. Taylor J H & Weisberg J M. Further experimental tests of relativistic gravity using the binary pulsar PSR 1913+16. Astrophys. J. 345:434-50. 1989.

Received January 4, 1993