Duncan Galloway (Monash University, Australia)

Thermonuclear burst spectroscopy with LOFT

Thermonuclear (type-I) X-ray bursts, caused by unstable ignition of accreted fuel on the surface of neutron
stars, provide insights in to nuclear burning processes as well as serving as probes of the interior conditions
of these objects. Understanding the shape of the emergent spectrum during bursts is critical to both of these
objectives. While the X-ray spectrum measured with various instruments is generally consistent with a
blackbody, at least two classes of exceptions are presently known.

First, the presence of significant residuals in extreme radius-expansion (ʻsuperexpansionʼ) bursts suggest the
presence of absorption edges, perhaps arising from nuclear burning ashes mixed into the radiation-driven
ejecta powered by the burst flux. Such features have been predicted theoretically, and can in principle
provide information on the neutron star compactness. However, high-quality spectra have proved difficult to
obtain, principally because such bursts are infrequent and unpredictable. The strength of such features in
weaker, more typical radius-expansion bursts is presently uncertain. Second, in a large sample of lowresolution
spectra observed by RXTE, we find a significant fraction that feature a hard excess compared to a
blackbody. This sample suggests a more complex underlying shape for the emergent spectrum, although
with the moderate signal to noise spectra currently available, it is difficult to further constrain the spectral
shape.

The high sensitivity and good spectral resolution goals for LOFT will likely provide significant new insights
into both these areas. Moderate exposure to superexpansion burst sources over the mission lifetime will
offer a high probability of detecting superexpansion bursts and confirming the nature and properties of the
putative absorption features. Observations of more typical radius-expansion bursts, which are weaker but
much more frequent, will permit searches for comparable features in those bursts. Similarly, higher 
signal-to-noise burst spectra will provide a more stringent test of the blackbody model, and likely 
reveal in detail the nature of the high-energy excess.