Mar 25, 1998

Scientist finds 2-in-1 burster; Pulsar goes off twice each orbit

click to navigate to topic

Scientist finds 2-in-1 burster;

Pulsar goes off twice each orbit

dark companion to pulsar drawing
March 25, 1998: A NASA scientist has found a new puzzle in the sky, an X-ray pulsar that appears to be in a lopsided orbit that makes it burst twice every "year" rather than once.

In a paper accepted for publication in The Astrophysical Journal, Colleen Wilson of NASA's Marshall Space Flight Center describes the discovery of an accreting X-ray pulsar which has no visible component.

When it is active, GRO J2058+42 appears to burst in X-rays twice each time it circles its primary star. (Above, right is an artist's concept of a pulsar/dark companion binary. Click on the image to see data plots showing BATSE-observed brightenings of a number of pulsars, including GRO J2058+42.)

"It's an unusual behavior," she said. "It's not like what we see with other accretion-powered pulsars." These are pulsars that burst in X-rays and gamma rays as they gobble gas in a disk from a larger parent star.

For more about NASA/Marshall's studies of X-ray pulsars with BATSE - including a detailed description of accreting pulsars and how they differ from radio pulsars - check our story about Things that go bump in the night.

Wilson made her discovery with the Burst and Transient Source Experiment aboard the Compton Gamma Ray Observatory, and built on it with additional observations by the Rossi X-ray Timing Explorer. This one of 12 known transient accreting X-ray pulsars with no visible companion.

Pulsars - rotating neutron stars - are among the most intriguing objects in the sky. They were found in 1965 when radio astronomers discovered several objects that emitted radio waves with clock-like precision. The sources soon were identified as rapidly rotating neutron stars with intense magnetic fields. Where radio pulsars have the regularity of a Swiss watch, accretion pulsars are like cheap alarm clocks that easily gain and lose time - and go off when you least expect it.

Since the April 1991 launch of BATSE, a NASA/Marshall instrument, astrophysicists have detected 20 of 45 known pulsars and discovered another 5 new X-ray pulsars. These accreting pulsars are not the same as the mysterious gamma-ray bursts that appear to be coming from the edge of the universe. They are energetic enough to be recorded by BATSE, but don't always get noticed right away.

So, as Wilson reviewed BATSE data in September 1995, she found a burst that registered 140 milliCrabs (or, 140/1,000ths the brightness of the Crab Nebula, which astrophysicists use as a standard candle). By using a computer to fold the data on itself, Wilson found that the source repeated every 198 seconds, an indication of a massive, compact object spinning at high speed.

At right is a plot of the pulse profile from 2058+42. The span from peak to peak is 198 seconds. (Links to


"The way we can tell it was something new is that the ones that aren't real don't repeat from day to day," Wilson said.

GRO J2058+42 (the numbers indicate its approximate position, 20 hrs, 58 minutes along the celestial equator, and 42 degrees north) had signaled the start of a giant burst spanning a 46-day period. The burst peaked at 140 milliCrab (140/1,000ths the brightness of the Crab Nebula). BATSE saw another 500 days of weaker bursts.

BATSE and the Rossi all-sky monitor showed bursts every 54 days, but BATSE saw the bursts alternate in brightness in a 110-day cycle. Clearly, Wilson was on to something new.

In the other accretion-powered pulsars, the emissions come only once an orbit when they do occur; sometimes they fade for weeks or months. J2058+42's twice-an-orbit bursts are new.

Wilson said one possibility is that J2058+42 is a binary star system composed of a type Be star (a type B star with emission lines), about 8 to 15 times the mass of our sun, with a neutron star in a lopsided orbit (below).


An excretion disk of gas, expelled by the Be star in an unusual stellar wind, orbits the Be star about even with the equator. But the neutron star is in an elliptical orbit inclined to the equator, so it sails in and out of the excretion disk twice every 110 days.

As a result, J2058+42 is quiet most of the time, and emits X-rays twice every orbit as the dense, magnetically intense neutron star plows through the cloud of gas. When the pulsar is near periastron (closest approach to the star), the bursts are more intense, now around 1.4 percent the brightness of the Crab Nebula. When the pulsar is near apastron (most distant from the star), the bursts are weaker, down to less than 1 percent of the Crab, almost at the threshold of the BATSE.

A more likely explanation, Wilson said, is that somehow the pulsar winds up the accretion disk at periastron and accelerates its consumption of matter, and at apastron it's taking a lighter meal from the stellar wind.

The identification is based on J2058+42's behavior matching that of other Be-pulsar binaries.

"About half the known pulsars in high-mass X-ray binaries are thought to be with Be stars," she said. The high mass here refers to the mass of the companion star, not the pulsar.

Left: The position of GRO J2058+42 is known to within 4 arc-min based on its detection by BATSE (red diamond), and scans by Compton's OSSE (yellow diamond) and scans by Rossi's proportional counter array instrument (links to

. Credit Colleen Wilson, NASA/Marshall Space Flight Center).

Unfortunately, no one has been able to get a look at the object for other observations that might confirm an optical identification. Wilson said this is the sixth time BATSE has found such an object for which no visible component can be found.

"Part of the problem is that the error box is so big," she said. Because BATSE is designed to watch the entire sky, it cannot locate a source with the same precision of a telescope zeroing in on a star. Other data from before September 1995 from BASTE and from Roentgensatellit (a European Space Agency satellite) also don't show anything that stands out as a possible source. Neither does a survey of optical images, possibly because interstellar dust simply dims the star before its light arrives here.

Even though observations with Rossi have narrowed the position estimate to 4 arc-minutes (about 1/8th the apparent diameter of the Moon), the error box is still quite large for an optical telescope to search. Wilson estimates the distance to J2058+42 at about 23,000 to almost 50,000 light years (almost a quarter of the way across the galaxy).

Meanwhile, J2058+42 has become too quiet for BATSE to detect. Wilson will keep checking the data, and also looks forward to observations from the Rossi Explorer which should be made this week.

Discovery of the 198 s X-ray Pulsar GRO J2058+42. Colleen A. Wilson, Mark H. Finger, B. Alan Harmon, Deepto Chakrabarty, and Tod Strohmayer. Astrophysical Journal (scheduled for Vol 499, June 1, 1998). Postscript copies of the preprints are available (422K file size).

Join our growing list of subscribers - sign up for our express news delivery and you will receive a mail message every time we post a new story!!!


Author: Dave Dooling
Curator: Bryan Walls
NASA Official: John M. Horack