Radius-expansion burst spectra from 4U 1728-34: an ultracompact binary?

Recent theoretical and observational studies have shown that ashes from thermonuclear burning may be ejected during radius-expansion bursts, giving rise to photoionisation edges in the X-ray spectra. We report a search for such features in Chandra spectra observed from the low-mass X-ray binary 4U 1728-34. We analysed the spectra from four radius-expansion bursts detected in 2006 July, and two in 2002 March, but found no evidence for discrete features. We estimate upper limits for the equivalent widths of edges of a few hundred eV, which for the moderate temperatures observed during the bursts, are comparable with the predictions. During the 2006 July observation 4U 1728-34 exhibited weak, unusually frequent bursts (separated by <2 hr in some cases), with profiles and alpha-values characteristic of hydrogen-poor fuel. Recurrence times as short as those measured are insufficient to exhaust the accreted hydrogen at solar composition, suggesting that the source accretes hydrogen deficient fuel, for example from an evolved donor. The detection for the first time of a 10.77 min periodic signal in the persistent intensity, perhaps arising from orbital modulation, supports this explanation, and suggests that this system is an ultracompact binary similar to 4U 1820-30.

💡 Research Summary

This paper investigates whether the ashes of thermonuclear burning, expected to be expelled during radius‑expansion bursts (REBs), leave detectable photo‑ionisation edges in the X‑ray spectra of the low‑mass X‑ray binary 4U 1728‑34. Using Chandra/HETGS data, the authors analysed six REBs—four observed in July 2006 and two in March 2002—by extracting time‑resolved spectra (0.5–2 s intervals) and fitting each with a continuum consisting of a blackbody plus a Comptonised component. They then added absorption‑edge components at energies corresponding to Fe‑peak and lighter elements and evaluated the statistical improvement. No edge was found; the 90 % confidence upper limits on equivalent widths lie between 200 and 500 eV, comparable to theoretical predictions for the moderate burst temperatures (≈2–3 keV).

The 2006 bursts displayed unusually short recurrence times (as brief as <2 h) and α‑values of 50–70, indicative of hydrogen‑poor fuel. Such rapid recurrence cannot be reconciled with the depletion of solar‑composition hydrogen, implying that the accreted material is already hydrogen‑deficient, as would be expected from an evolved donor (e.g., a helium white dwarf).

A key ancillary result is the detection of a persistent‑flux modulation with a period of 10.77 min (≈0.00155 Hz) and an amplitude of ~1 %. The stability of this signal across the observation suggests an orbital origin, placing 4U 1728‑34 among the class of ultracompact binaries (UCXBs) whose orbital periods are typically <80 min. The period is strikingly similar to that of the well‑studied UCXB 4U 1820‑30, reinforcing the hypothesis that 4U 1728‑34 harbours a compact, hydrogen‑deficient donor.

The authors discuss two possible reasons for the non‑detection of edges. First, the actual edge strength may be intrinsically weaker than predicted because the burst atmosphere’s temperature and density structure limit the ionisation of heavy elements. Second, the limited sensitivity and time resolution of the current data may miss very narrow, transient edges that appear only for fractions of a second. They argue that future missions with higher effective area and finer temporal resolution (e.g., NICER, XRISM, Athena) will be essential to probe these subtle features.

In summary, the study provides three major contributions: (1) a stringent observational upper limit on burst‑induced photo‑ionisation edges that is consistent with existing theoretical models; (2) evidence from burst recurrence times and α‑values that the accreted fuel is hydrogen‑poor, supporting an ultracompact binary scenario; and (3) the first detection of a ~10.8 min periodic modulation in the persistent emission of 4U 1728‑34, likely representing the orbital period. The work underscores the importance of combining high‑resolution spectroscopy with timing analysis to unravel the nature of accretion and nuclear burning in compact binaries, and it sets the stage for more sensitive future observations that could finally capture the elusive spectral signatures of burst ashes.