Long term variability of Cygnus X-1: VII. Orbital variability of the focussed wind in Cyg X-1 / HDE 226868 system

TL;DR

This study investigates the orbital variability of the focussed wind in the Cyg X-1 system using 16 years of X-ray data, revealing complex absorption patterns influenced by wind clumping and state-dependent effects.

Contribution

It provides a detailed analysis of wind variability in Cyg X-1 and compares observations with simplified models, highlighting the importance of wind clumping and focusing effects.

Findings

Absorption varies strongly over the orbital period and shows large scatter.

Orbital variability is most prominent in the black hole's hard X-ray state.

Simplified clumpy wind models show good qualitative agreement with observations.

Abstract

Binary systems with an accreting compact object are a unique chance to investigate the strong, clumpy, line-driven winds of early type supergiants by using the compact object's X-rays to probe the wind structure. We analyze the two-component wind of HDE 226868, the O9.7Iab giant companion of the black hole Cyg X-1 using 4.77 Ms of RXTE observations of the system taken over the course of 16 years. Absorption changes strongly over the 5.6 d binary orbit, but also shows a large scatter at a given orbital phase, especially at superior conjunction. The orbital variability is most prominent when the black hole is in the hard X-ray state. Our data are poorer for the intermediate and soft state, but show signs for orbital variability of the absorption column in the intermediate state. We quantitatively compare the data in the hard state to a toy model of a focussed Castor-Abbott-Klein-wind: as it does not incorporate clumping, the model does not describe the observations well. A qualitative comparison to a simplified simulation of clumpy winds with spherical clumps shows good agreement in the distribution of the equivalent hydrogen column density for models with a porosity length on the order of the stellar radius at inferior conjunction; we conjecture that the deviations between data and model at superior conjunction could be either due to lack of a focussed wind component in the model or a more complicated clump structure.