Skip to main content
SearchLogin or Signup

Spectral Modeling of State-Transitioning Disk-Wind in BH XRBs

Presentation #524.07 in the session “Black Holes”.

Published onJan 11, 2021
Spectral Modeling of State-Transitioning Disk-Wind in BH XRBs

An apparent feature distinct to BH XRBs, discovered by our ability to follow them through entire cycles of flux change, is a hysteresis in their spectral/timing properties, which follow a well-known q-diagram. This trend outlines the great diversity of the spectral/timing properties, as well as the global evolution of accretion disk properties with their accretion rate. The diagram generally exhibits two phenomenological regions from spectroscopic observations; high/soft (“wind-on”) state where (blueshifted) ionized absorbers of v < 1,000 km/s are detected and low/hard (“wind-off”) state in the apparent absence of such absorbers, thus perhaps implying a wind dichotomy. Motivated with the state-of-the-art Chandra grating data from GRO J1655-40, H1743-322 and 4U 1630-47 as exemplary sources, we present a disk-wind model in an effort to systematically understand the underlying X-ray wind condition in both states in the context of a magnetic driving scenario. By utilizing a multi-ion spectral fitting for a broad-band spectrum (e.g. 1-10 keV) within the framework of 3D magnetized disk winds, we explicitly demonstrate how wind physical properties should change across the state transition with a special focus on wind density and density gradient of a global wind structure. Our spectral modeling strongly supports the idea that the internal structure of the wind (in addition to the ionizing X-ray SED) must change to explain the observed multi-epoch data and the wind is persistently present all the time, while spectroscopically invisible during “wind-off” state. In anticipation of the upcoming new X-ray micro-calorimeter instruments such as XRISM/Resolve and Athena/X-IFU, we also present a preliminary simulation of the expected absorption spectra in both states showing a feasibility of testing the model.


No comments here