relxillNS: Neutron-Star X-ray Reflection Model

• relxillNS is a relativistic X-ray reflection model that adapts blackbody illumination for neutron-star accretion disks rather than the standard power-law continuum.
• It combines local reflection computations with Kerr-metric based relativistic transfer to capture disk parameters like ionization, electron density, and emissivity profiles.
• Widely applied in modeling Fe K emissions, Compton humps, and burst-driven disk reprocessing, it is a key tool for analyzing neutron-star low-mass X-ray binaries.

Searching arXiv for relxillNS and recent applications to ground the article in current literature. relxillNS is a relativistic X-ray reflection model for accreting neutron stars in which the primary illuminating continuum is a single-temperature blackbody emitted either at the surface of the star or at the boundary layer, rather than the power-law continuum usually assumed for accreting black holes (Garcia et al., 2021). It is the relativistic counterpart of xillverNS and is distributed as a neutron-star-specific flavor within the relxill suite (Garcia et al., 2021). In practice, relxillNS has become a standard tool for modeling Fe K emission, Compton humps, burst-driven disk reprocessing, and inner-disk geometry in neutron-star low-mass X-ray binaries, including Aql X-1, 4U 1636–536, Cygnus X-2, Serpens X-1, Sco X-1, GX 13+1, GX 340+0, Swift J17480-2446, and 4U 1820–30 (Mandal et al., 22 Mar 2025).

1. Definition and lineage

relxillNS was introduced together with xillverNS as a set of reflection models “specifically tailored to model the X-ray radiation reprocessed in accretion disks around neutron stars,” under the assumption that the irradiating spectrum is blackbody-like and originates in the neutron-star surface or boundary layer (Garcia et al., 2021). This point is central: the model is not merely a reparameterized black-hole reflection model, but a different illumination prescription designed for neutron-star accretion physics.

The model inherits the broader relxill strategy of combining a local reflection calculation with relativistic transfer, but its local spectrum is computed for blackbody illumination rather than for a power-law or cutoff power-law (Garcia et al., 2021). Earlier blackbody-illuminated reflection calculations existed, including BBrefl and reflionx_BB, but relxillNS was presented as an updated, publicly available implementation with relativistic blurring and parameter coverage appropriate to neutron-star disks (Garcia et al., 2021).

The Kerr metric is used for the relativistic transfer. For neutron stars, the stated justification is that for low spin the Kerr approximation remains adequate, with discrepancies of order $\le 10\%$ at $a_*<0.3$ (Garcia et al., 2021). This makes relxillNS most naturally suited to slowly rotating or moderately rotating neutron-star systems, while preserving the standard relativistic observables used in disk-reflection work: inclination, inner radius, emissivity profile, ionization state, and elemental abundance.

2. Physical assumptions and parameterization

The underlying geometry is a geometrically thin, optically thick accretion disk modeled locally as a plane-parallel slab of constant density, illuminated by a single-temperature blackbody (Garcia et al., 2021). The ionization state is parameterized by

$\xi = \frac{4\pi F_{\rm x}}{n_e},$

where $F_{\rm x}$ is the incident flux and $n_e$ is the electron density (Garcia et al., 2021). Radial illumination is typically described by an emissivity law of the form $\epsilon(r)\propto r^{-q}$, either as a single power law or a broken power law depending on the application (Garcia et al., 2021).

The model space reported for relxillNS includes the following principal ranges (Garcia et al., 2021):

Parameter	Meaning	Range
$kT_{\rm bb}$	Blackbody temperature	$0.5$–$10$ keV
$\log \xi$	Ionization parameter	$a_*<0.3$0–$a_*<0.3$1
$a_*<0.3$2	Electron density	$a_*<0.3$3–$a_*<0.3$4
$a_*<0.3$5	Iron abundance	$a_*<0.3$6–$a_*<0.3$7
$a_*<0.3$8	Spin parameter	$a_*<0.3$9 to $\xi = \frac{4\pi F_{\rm x}}{n_e},$0
$\xi = \frac{4\pi F_{\rm x}}{n_e},$1	Inclination	$\xi = \frac{4\pi F_{\rm x}}{n_e},$2–$\xi = \frac{4\pi F_{\rm x}}{n_e},$3
$\xi = \frac{4\pi F_{\rm x}}{n_e},$4	Inner disk radius	$\xi = \frac{4\pi F_{\rm x}}{n_e},$5–$\xi = \frac{4\pi F_{\rm x}}{n_e},$6 ISCO
$\xi = \frac{4\pi F_{\rm x}}{n_e},$7	Outer disk radius	$\xi = \frac{4\pi F_{\rm x}}{n_e},$8–$\xi = \frac{4\pi F_{\rm x}}{n_e},$9

In many applied fits, several of these parameters are fixed because burst or time-resolved