The first thermonuclear X-ray burst from IGR J00291+5934
ATel #7852; E. Bozzo (ISDC, Switzerland), A. Papitto (ICE CSIC-IEEC, Spain), A. Sanna (UNICA, Italy), M. Falanga (ISSI, Switzerland), S. Campana (INAF-OA Brera, Italy), T. di Salvo, R. Iaria (Univ. of Palermo, Italy), F. Pintore, A. Riggio (UNICA, Italy), E. Egron (INAF-OA Cagliari, Italy)
on 28 Jul 2015; 20:47 UT
Credential Certification: E. Bozzo (firstname.lastname@example.org)
Subjects: X-ray, Neutron Star, Transient, Pulsar
Referred to by ATel #: 7949
We performed a time resolved spectral analysis of the Swift/XRT observation in WT mode (ID 00650221000) of the accreting millisecond X-ray pulsar (AMXP) IGR J00291+5934 (Atel #7836, #7837), recently reported by Kuin et al. (Atel #7849). As mentioned by these authors, this is the first thermonuclear burst recorded from the source.
We extracted 9 spectra during the XRT observation, with durations lasting from 10 to 50 seconds (depending on the recorded source count-rate). We used the source spectrum extracted from first 60 seconds of the WT observation performed about 4 ks after the burst (ID 00650221001) as background in the fits to the time resolved burst spectra.
All spectra could be well fit with an absorbed blackbody model (chi^2_red=1.05/887 d.o.f.). The absorption column density was forced to be the same for all spectra in the fit. We obtained N_H = (0.25+/-0.04)E22 cm^(-2) and measured a progressive decay of the blackbody temperature from 1.8+/-0.1 keV to 1.01+/-0.06 keV. The corresponding observed flux in the 0.5-10 keV energy range decreased from (1.3+/-0.1)E-8 erg/cm^2/s to (1.4+/-0.2)E-9 erg/cm^2/s. The blackbody radius remained roughly constant for all spectra at about 5 km (assuming a distance of 4 kpc; see, e.g., Torres et al. 2008, ApJ, 672, 1079).
The decay of the blackbody temperature supports the interpretation of the event as the tail of a thermonuclear, type-I X-ray burst.
As the XRT started observing the source about 119 s after the BAT trigger (Atel #7849
), it is likely that the event observed by XRT and BAT are related. Following such assumption, the thermonuclear burst observed from IGR J00291+5934 would have lasted for about 400 s in total. In such a case we suggest that the thermonuclear burst was ignited in a pure He layer, after the steady H-burning exhausted the accreted hydrogen at the base of the layer. A photospheric radius expansion (PRE) might have occurred before XRT started observing the source. Unfortunately, given the lack of XRT data during the onset of the event, it is not possible to carry out a more refined analysis.
Thermonuclear explosions similar to that observed from IGR J00291+5934 last longer compared to the other more common type-I bursts observed from the same class of objects, indicating a much longer recurrence time and thus a lower probability of being observed during the relatively short outbursts of AMXPs (~a week).
After the discovery of a type-I X-ray burst from IGR J00291+5934, at least one X-ray burst has been observed from all the accreting millisecond pulsars with H-rich donors and orbital periods comprised between 2.01 and 11 hr. So far, no X-ray bursts have been detected from AMXPs in ultracompact binaries (orbital periods <40 min).