Investigation of X-ray timing and spectral properties of ESO 243-49 HLX-1 with long-term Swift Monitoring

TL;DR

This study analyzes long-term Swift data of ESO 243-49 HLX-1, revealing increasing outburst intervals, spectral state changes, and exploring models for its quasi-periodic outbursts, suggesting a shorter true orbital period.

Contribution

It provides detailed timing and spectral analysis of HLX-1, confirming increasing outburst intervals and evaluating different physical scenarios for its variability.

Findings

Outburst intervals are increasing over time.

Spectral analysis shows a power-law in quiescence and thermal dominance during outbursts.

Models suggest the true orbital period is shorter than observed quasi-periodicity.

Abstract

The long-term Swift monitoring of ESO 243-49 HLX-1 provides an opportunity to investigate the detailed timing and spectral behaviour of this hyper-luminous X-ray source. Swift has detected 7 outbursts since 2009 mid-August. Using different dynamical timing algorithms, we confirm an increasing trend for the time intervals between outbursts, which is manifest in the delays between the latest outbursts. The X-ray spectra of HLX-1 in quiescence can be described with a single power-law model while the thermal component dominates the X-ray emission during outburst. There is only marginal evidence for photon index (or spectral hardness) changes between quiescent states with about 1{\sigma} deviation. With the updated temporal and spectral features, we re-examine different scenarios to explain the origin of the quasi-periodic modulation of HLX-1. A significantly increasing trend without obvious stochastic fluctuations on the timescale of the detected quasi-period may not fully support an orbital period origin as might be due to mass transfer episodes from a donor star at periastron of an extremely eccentric orbit. The outburst profile seems to be consistent with the effect of tidal-induced-precession of an accretion disc or an oscillating wind scenario in the inner disc. Based on these models, we speculate that the true orbital period is much shorter than the detected quasi-periodicity.