Jets from black hole X-ray binaries: testing, refining and extending empirical models for the coupling to X-rays

TL;DR

This study analyzes radio and X-ray data from black hole X-ray binaries to test and refine a model linking accretion states to jet ejections, confirming some aspects while highlighting areas needing further investigation.

Contribution

It extends the unified model by incorporating timing properties and analyzing a large sample of outbursts, providing new insights into jet ejection timing and state transitions.

Findings

Radio peaks occur near hard-to-soft state transitions.

Radio flaring correlates with X-ray flaring during transitions.

Soft states show suppressed or relic radio emission.

Abstract

In this paper we study the relation of radio emission to X-ray spectral and variability properties for a large sample of black hole X-ray binary systems. This is done to test, refine and extend -- notably into the timing properties -- the previously published `unified model' for the coupling of accretion and ejection in such sources. In 14 outbursts from 11 different sources we find that in every case the peak radio flux, on occasion directly resolved into discrete relativistic ejections, is associated with the bright hard to soft state transition near the peak of the outburst. We also note the association of the radio flaring with periods of X-ray flaring during this transition in most, but not all, of the systems. In the soft state, radio emission is in nearly all cases either undetectable or optically thin, consistent with the suppression of the core jet in these states and `relic' radio emission from interactions of previously ejected material and the ambient medium. However, these data cannot rule out an intermittent, optically thin, jet in the soft state. In attempting to associate X-ray timing properties with the ejection events we find a close, but not exact, correspondence between phases of very low integrated X-ray variability and such ejections. In summary we find no strong evidence against the originally proposed model, confirming and extending some aspects of it with a much larger sample, but note that several aspects remain poorly tested. (ABRIDGED)