Geometrical beaming of stellar mass ULXs

TL;DR

This paper investigates the geometry of ultraluminous X-ray sources (ULXs) and finds that their observed variability suggests a conical outflow and geometrical beaming, challenging the assumption of isotropic emission from stellar-mass black holes.

Contribution

It introduces a method to relate eclipse-induced variability to system geometry and demonstrates that most ULXs are viewed at angles consistent with beaming from super-critical accretion flows.

Findings

Most ULXs have variability too low for eclipses, indicating non-isotropic viewing angles.

The data supports the presence of conical outflows and geometrical beaming in ULXs.

Eclipses are unlikely in the majority of observed ULXs, implying beamed emission.

Abstract

The presence or lack of eclipses in the X-ray lightcurves of ultraluminous X-ray sources (ULXs) can be directly linked to the accreting system geometry. In the case where the compact object is stellar mass and radiates isotropically, we should expect eclipses by a main-sequence to sub-giant secondary star on the recurrence timescale of hours to days. X-ray lightcurves are now available for large numbers of ULXs as a result of the latest XMM-Newton catalogue. We determine the amount of fractional variability that should be injected into an otherwise featureless lightcurve for a given set of system parameters as a result of eclipses and compare this to the available data. We find that the vast majority of sources for which the variability has been measured to be non-zero and for which available observations meet the criteria for eclipse searches, have fractional variabilities which are too low to derive from eclipses and so must be viewed such that theta =< acos(R*/a). This would require that the disc subtends a larger angle than that of the secondary star and is therefore consistent with a conical outflow formed from super-critical accretion rates and implies some level of geometrical beaming in ULXs.