The direct cooling tail method for X-ray burst analysis to constrain neutron star masses and radii

TL;DR

This paper introduces an improved direct cooling tail method for analyzing X-ray bursts from neutron stars, enabling more accurate constraints on their masses and radii by fitting spectral data directly to theoretical models.

Contribution

The paper develops a generalized, direct fitting approach that accounts for bolometric corrections and fits data directly on the M-R plane, improving neutron star parameter estimation.

Findings

Confidence regions shift towards larger radii compared to standard methods.

Estimated neutron star radius: 11.5-13.0 km for mass 1.3-1.8 M_sun.

Method demonstrated on SAX J1810.8--2609 data.

Abstract

Determining neutron star (NS) radii and masses can help to understand the properties of matter at supra-nuclear densities. Thermal emission during thermonuclear X-ray bursts from NSs in low-mass X-ray binaries provides a unique opportunity to study NS parameters, because of the high fluxes, large luminosity variations and the related changes in the spectral properties. The standard cooling tail method uses hot NS atmosphere models to convert the observed spectral evolution during cooling stages of X-ray bursts to the Eddington flux F_Edd and the stellar angular size \Omega. These are then translated to the constraints on the NS mass M and radius R. Here we present the improved, direct cooling tail method that generalises the standard approach. First, we adjust the cooling tail method to account for the bolometric correction to the flux. Then, we fit the observed dependence of the blackbody normalization on flux with a theoretical model directly on the M-R plane by interpolating theoretical dependences to a given gravity, hence ensuring only weakly informative priors for M and R instead of F_Edd and \Omega. The direct cooling method is demonstrated using a photospheric radius expansion burst from SAX J1810.8--2609, which has happened when the system was in the hard state. Comparing to the standard cooling tail method, the confidence regions are shifted by 1 sigma towards larger radii, giving R=11.5-13.0 km at M=1.3-1.8 M_sun for this NS.