The Astronomer's Telegram
Telegram Index | Search
To The Parent Site

[ Previous | Next ]

The Eclipsing X-ray Burster in Terzan 5 is Probably Not the 2000 Transient

ATEL # 2933; C. O. Heinke (Alberta), M. Linares (MIT), J. Gladstone (Alberta), J. Homan (MIT), D. Altamirano (Amsterdam), D. Pooley (Eureka Scientific)
on 13 Oct 2010; 19:23 UT
Password Certification: Craig Heinke (

Subjects: X-ray, Binaries, Globular Clusters, Neutron Stars, Pulsars, Transients
Referred to by ATEL #: 2935, 2937, 2939, 2940, 2946, 2952, 2958, 2974, 3000, 3044

The transient LMXB and 11 Hz pulsar currently in outburst in Terzan 5 (ATEL #2919, #2920, #2922, #2924, #2932) has been recently observed to show at least one eclipse of 511 s duration (ATEL #2929). We have checked the Chandra lightcurves (readout streak and dust-scattered halo) from the 2000 July 24 observation (Heinke et al. 2003, ApJ, 590, 809), and find no eclipses in the 45 ks (12.5 hour) observation. Thus if the transient is the same X-ray source, its period is >12.5 hours. Such a system must have a subgiant donor star, of mass ~0.9 solar masses. Assuming a 1.4 solar mass neutron star, R_2/a=0.34 (Frank, King & Raine 1992), and the maximum eclipse length would be 1.35 hours. To produce an eclipse of only 511 seconds requires an extremely grazing eclipse geometry, with a chance probability (given the fact of eclipses) of only 0.5%. Thus we conclude that it is unlikely that the currently outbursting LMXB in Terzan 5 is the same transient as observed in 2000 (CXOGLB J174805.2-244647). This conclusion is strengthened by the lack of eclipses observed in multiple RXTE observations of the 2000 outburst.

The published Swift error circle (ATEL #2920) suffers from incomplete sampling of the PSF (on the edge of the chip). A second Swift observation (00031841001) gives a different centroid, RA=267.019969, Dec=-24.77968, with (90% conf. error) 3.56 arcseconds, which is affected by a readout streak removing the center of the PSF. Each error circle includes multiple X-ray sources; the new Swift error circle includes the candidate quiescent LMXBs CX25 and CX9 from Heinke et al. 2006 (ApJ, 651, 1098), along with CX19 and CX28, while the old error circle includes candidate quiescent LMXB CX18 along with CX35. A Chandra observation has been approved, and we encourage further observations. We thank NSERC for support.

[ Telegram Index ]

R. E. Rutledge , Editor
ATEL Mirror v1.0 Updates