Binary Pulsars and Evidence for Gravitational Radiation Matthew S. Paoletti Institute for Research in Electronics and Applied Physics Energy Research Facility, Bldg. #223 University of Maryland College Park, MD 20742-3511 ABSTRACT A discussion of binary pulsars and their utility in providing evidence for gravitational radiation is presented. The results focus on the binary pulsar PSR 1913+16, as it was the first discovered and most studied. Assuming general relativity is correct, the masses of the pulsar and its companion are determined to be and times the mass of the Sun. The decay rate of the orbital period due to gravitational radiation emission is measured to be 1.0013 ± 0.0021 times the general relativistic prediction, providing strong evidence for the existence and nature of gravitational radiation. 11.4414 0.0002m =±21.3867 0.0002m =± 1. INTRODUCTION Prior to the discovery of binary pulsars all tests of general relativity and other theories of gravity were restricted to the weak-field, slow-motion interactions present within the solar system. In this limit nonlinearities and the predicted effects of gravitational radiation are negligible. Therefore it was greatly desired to find or create a system sufficiently relativistic to extend the experimental tests of general relativity and other competing theories of gravity. These tests were made possible in 1975 with the discovery of a binary pulsar system by Hulse and Taylor1. Binary pulsars are composed of at least one pulsar and a companion massive object. Pulsars regularly emit detectable pulses of radiation and thereby serve as extremely stable clocks. The masses of the two bodies are often on the order of the solar mass and they rapidly orbit each other. By virtue of their large mass and rapid orbits, binary pulsars provide the experimentalist the opportunity to test the predictions of general relativity in the strong-field, rapid motion limit. Consequently, great effort has been put forth to discover and characterize many pulsar systems. Figure 1 shows a celestial map2 of all known pulsars as of July 1994. Fig. 1. Distribution of 558 pulsars in Galactic coordinates. The center of the galaxy is in the middle, and longitude increases to the left. One of the fundamental predictions of general relativity is the emission of quadrupolar gravitational radiation. However, being a relativistic effect, gravitational radiation emission is only significant in the relativistic limits of strong gravitational fields and rapid accelerations. These limits do not exist in the solar systemand therefore gravitational radiation was undetectable within the confines of the solar system. Binary pulsars, on the other hand, are sufficieFig. 2. Spin-down rates and periods of pulsars The decay of the orbital period of a binary gravitational radiation emission. Figure 2 2. DISCOVERY OF A BINARY PULSAR The covered by urnell and Hewish in 1967. The research team wematic survey of pulsars nsued at the Arecibo Observatory in Puerto Rico lntly relativistic that gravitational radiation emission is expected to be significant. Gravitational radiation has yet to be detected directly. However, gravitational radiation carries both energy and angular momentum away from the orbiting bodies. These losses produce a decrease in the orbital period and eccentricity of the binary system. By observing time variations in the pulsar period it is possible to determine both the slow-down rate of the orbit and the change in eccentricity3, , ,4 5 6. observed before July 1994. Pulsars known to be in binary systems are denoted by larger circles around the dots. Symbols aligned near the bottom represent pulsars for which the orbital period decay has yet to be measured. pulsar system may serve as an indirect test of theoretical predictions for illustrates the observed spin-down rate and period of a collection of pulsars2. Binary systems are marked by larger circles around dots. The spin-down rate of binaries spans nine orders of magnitude, providing unique opportunities to test gravity theories. first pulsar was dis7Bas investigating the scintillation of quasars using a radio array when they found a regular signal of pulsed radiation with a period of roughly a few seconds. The team first sought to prove the origin of the signal was not terrestrial. This was achieved by showing that the radiation source reappeared every sidereal day rather than every solar day. Due to the abnormally regular signal the team dubbed the source LGM-1 for “little green men,” as a comical reference to extraterrestrial life. After it was determined that the pulsar was a rapidly rotating neutron star it was named PSR 1919+21. A systeeading to forty well-characterized pulsars prior to the discovery of PSR 1913+16 by Hulse and Taylor1. The 59-ms pulsar was first detected in July 1974. The group sought to determine the period of the pulsar to high precision as they had for others. Their attempts to measure the period with a precision of 1 sμ± were impeded by apparent changes in the period of up to 80 sμ from day to day. At times the period would change by as much as 8 sμ over a 5 e observation. This apparent change in period was atypical of these seemingly “perfect clocks.” The largest observed changes in period of other pulsars were on minutthe order of 10 sμper year, and typical changes were many orders of magnitude lower8. Upon investigation it was determby orb y curve as determined from Doppler binary pulsar PSR 16. Points pplying the above and all other measured rbital parameters it was concluded that the e first 3. DETERMINATION OF PULSAR AND COMPANION MASSES Through non-relativistic analysis of the time of arrival data it is possible to obtain furtherined that Doppler shifts produced ital motion of the pulsar correctly explicate the observed period changes. The Doppler shifts allowed the researchers to determine the radial velocity as a function of orbital phase as shown in figure 31. The period of the orbit was determined to be 27908 7 sbP =±and the eccentricity0.615 0.010e =±. Fig. 3. Velocitshifts for the 1913+represent measurements of the pulsar period distributed over parts of 10 different orbital periods. The curve represents a fit to the data as discussed in the reference. Aochange of the pulsar period in the center of mass frame is1210−< , similar to that of other pulsars.