[Paper Review] FERO (Finding Extreme Relativistic Objects): statistics of relativistic broad Fe K alpha lines in AGN

A. L. Longinotti, I. de la Calle Pérez+3 more|ArXiv.org|Sep 20, 2007

Astrophysical Phenomena and Observations1 references3 citations

TL;DR

This study analyzes 157 XMM-Newton observations of radio-quiet AGN to investigate relativistically broadened Fe K𝛼 lines, using refined spectral modeling and stacking techniques. It finds a 10% detection fraction for broad Fe K𝛼 lines at 5σ significance across the full sample, rising to 33% in a flux-limited sub-sample, with no significant correlation between line equivalent width and black hole mass, accretion rate, or luminosity.

ABSTRACT

The properties of the relativistically broadened Fe K alpha line emitted in Active Galactic Nuclei (AGN) are still debated among the AGN community. Recent works seem to exclude that the broad Fe line is a common feature of AGN. The analysis of a large sample composed by 157 XMM-Newton archival observations of radio quiet AGN is presented here. This ongoing project is a development of the work reported in Guainazzi et al. 2006.

Motivation & Objective

• To determine the detection frequency and statistical properties of relativistically broadened Fe K𝛼 lines in a large sample of radio-quiet AGN.
• To assess whether the broad Fe K𝛼 line is a common feature in AGN, resolving conflicting claims in prior literature.
• To investigate correlations between broad Fe K𝛼 line equivalent width and physical parameters such as black hole mass, accretion rate, and X-ray luminosity.
• To use stacked spectra of non-detections to probe the average Fe K𝛼 profile in under-exposed sources and constrain the presence of unresolved broad lines.

Proposed method

• Analyzes 157 XMM-Newton archival observations of radio-quiet AGN with N_H < 10^22.5 cm⁻².
• Employs a baseline spectral model including a power-law continuum, Compton reflection, narrow Fe K𝛼, Kβ, and Compton shoulder lines, and ionized absorption to model spectral curvature.
• Uses the kyrline model (Dovčiak et al. 2004) to fit relativistically broadened Fe K𝛼 emission from the accretion disc.
• Applies a 5σ significance threshold to define line detections and computes detection fractions across the full sample and a flux-limited sub-sample (F_2-10 > 1.8×10⁻¹¹ erg cm⁻² s⁻¹).
• Performs spectral stacking on sources with no significant broad line detection, using continuum-normalized ratios to search for residual broad line features.
• Splits stacked sources into luminosity bins to assess potential luminosity-dependent trends in broad line strength.

Experimental results

Research questions

• RQ1 What is the detection fraction of relativistically broadened Fe K𝛼 lines in a large, homogeneous sample of radio-quiet AGN?
• RQ2 Is there a significant correlation between the equivalent width of the broad Fe K𝛼 line and black hole mass or accretion rate?
• RQ3 Does the equivalent width of the broad Fe K𝛼 line correlate with the 2–10 keV X-ray luminosity of the AGN?
• RQ4 Can stacked spectra of non-detections reveal evidence for a weak or unresolved broad Fe K𝛼 line, and how does its strength vary with X-ray luminosity?

Key findings

• The broad Fe K𝛼 line detection fraction is 10% for the full sample of 157 AGN at a 5σ significance threshold.
• In a flux-limited sub-sample of 22 sources, the detection fraction increases to 33%, indicating that better-exposed sources are more likely to show the broad line.
• No significant correlation is observed between the broad Fe K𝛼 line equivalent width and black hole mass, accretion rate, or 2–10 keV luminosity across the sample.
• Stacked spectra of non-detections show no significant broad line profile, with residuals consistent with the continuum model below 6.4 keV.
• In luminosity-binned stacking, the broad line intensity is estimated to be at most 50 eV and appears to weaken with increasing X-ray luminosity.
• Theoretical broad line profiles do not match the stacked residuals, suggesting the broad line is either absent or too weak to be detected in the average spectrum of under-exposed sources.