Chandra verifies the current transient outburst in the globular cluster Terzan 5 is from Terzan 5 X-3, aka Swift J174805.3-244637

ATel #15953; C. O. Heinke (U. Alberta), J. Homan (Eureka Scientific), M. van den Berg (Harvard/CfA), S. Guillot (IRAP-CNRS), A. Bahramian (Curtin U.), D. Pooley (Trinity U.), G. Sivakoff (U. Alberta)
on 19 Mar 2023; 21:01 UT
Credential Certification: Craig Heinke (cheinke@virginia.edu)

Tweet

Negoro et al. (ATel #15917) reported a MAXI detection of a new outburst from the globular cluster Terzan 5, which contains multiple X-ray sources (e.g. Heinke et al. 2006, ApJ 651, 1098). The outburst was also seen by Integral (Atel #15921). At least three transient neutron star X-ray binaries have previously been identified in outburst in this cluster (e.g. ATels #2974, #3743, #4302, #7262). Swift observations (ATel #15919) indicated that the new transient may be associated with Terzan 5 X-3, aka Swift J174805.3-244637 (ATel #4302; Bahramian et al. 2014, ApJ 780, 127); however the 3.5" error radius of the Swift/XRT data also encompassed 5 other known X-ray sources in the cluster, including the redback pulsar Terzan 5 P (Urquhart et al. 2020, ApJ 904, 147). NICER observations (ATel #15922) also suggested that the X-ray bursts appeared significantly longer (100 s) than the one burst seen from Ter 5 X-3(τ=16 s, visible for 50 s; Bahramian et al. 2014).

We obtained a Chandra observation on Mar 19, 2023 for 10.6 ks. In addition to the bright transient with a readout streak, three cluster sources are detected by wavdetect within 1' of the cluster center, which we identify with detected sources CX4, CX5, and CX7 from Heinke et al. 2006. We estimate the position of the bright transient by eye, centering a circle on the piled-up annulus, and drawing a line through the middle of the readout streak. Boresighting the new observation using the three detected sources, we find a position for the transient of 267.02258(3), -24.77713(7), in the frame of Heinke et al. 2006; this is consistent with their position of CX2 (aka Ter 5 X-3), at 267.02255, -24.77713, with errors <0.3". We therefore identify the current transient as Ter 5 X-3 at high confidence.

We detected a single Type-I X-ray burst from the source, starting around at 15:33:35 UTC. Its total duration was ~50 seconds. This is considerably shorter than that of the two bursts detected with NICER during the early rising phase of the ongoing outburst (ATel #15922), but similar to that of a third burst detected with NICER during the current, more luminous phase.

The timing of this outburst (~11 years after Ter 5 X-3's 2012 outburst) matches the prediction of a 8-10 year recurrence time suggested by Bahramian et al. 2014 by comparing the quiescent luminosity of Ter X-3 with cooling calculations, and reinforces their suggestion that the 2002 outburst (Atel #101), which did not have a precise position, may have also been Ter 5 X-3.

We thank the Chandra X-ray Center for their quick response and assistance in scheduling this observation.