The Physics of the 'Heartbeat' State of GRS 1915+105

TL;DR

This study provides a detailed phase-resolved spectral analysis of the 'heartbeat' oscillation in GRS 1915+105, revealing the connection between X-ray variability, accretion disk wind, and jet activity, and supporting the radiation pressure instability model.

Contribution

First detailed spectral analysis linking X-ray oscillations with disk wind and jet activity in GRS 1915+105, demonstrating a causal disk-jet-wind connection.

Findings

X-ray spectrum changes affect disk wind structure within 5 seconds

Inner disk evaporation occurs during high-amplitude flares

Disk wind mass loss may drive long-term accretion oscillations

Abstract

We present the first detailed phase-resolved spectral analysis of a joint Chandra High Energy Transmission Grating Spectrometer and Rossi X-ray Timing Explorer observation of the rho variability class in the microquasar GRS 1915+105. The rho cycle displays a high-amplitude, double-peaked flare that recurs roughly every 50 s, and is sometimes referred to as the "heartbeat" oscillation. The spectral and timing properties of the oscillation are consistent with the radiation pressure instability and the evolution of a local Eddington limit in the inner disk. We exploit strong variations in the X-ray continuum, iron emission lines, and the accretion disk wind to probe the accretion geometry over nearly six orders of magnitude in distance from the black hole. At small scales (1-10 R_g), we detect a burst of bremsstrahlung emission that appears to occur when a portion of the inner accretion disk evaporates due to radiation pressure. Jet activity, as inferred from the appearance of a short X-ray hard state, seems to be limited to times near minimum luminosity, with a duty cycle of ~10%. On larger scales (1e5-1e6 R_g) we use detailed photoionization arguments to track the relationship between the fast X-ray variability and the accretion disk wind. For the first time, we are able to show that changes in the broadband X-ray spectrum produce changes in the structure and density of the accretion disk wind on timescales as short as 5 seconds. These results clearly establish a causal link between the X-ray oscillations and the disk wind and therefore support the existence of a disk-jet-wind connection. Furthermore, our analysis shows that the mass loss rate in the wind may be sufficient to cause long-term oscillations in the accretion rate, leading to state transitions in GRS 1915+105.