[Paper Review] Deep Chandra X-ray Observations of Low Mass X-ray Binary Candidates in the Early-Type Galaxy NGC 4697

G. R. Sivakoff, Andrés Jordán+3 more|ArXiv.org|Jun 3, 2008

Galaxies: Formation, Evolution, Phenomena2 references3 citations

TL;DR

This study uses deep Chandra X-ray observations and Hubble imaging of NGC 4697 to investigate the connection between low-mass X-ray binaries (LMXBs) and globular clusters (GCs), and to characterize the LMXB luminosity function (LF). It finds that LMXB occurrence in GCs scales with encounter rate (Γₕ)⁰.⁷⁹⁺⁰.¹⁸₋₀.¹⁵ and metallicity (Z/Z☉)⁰.⁵⁰⁺⁰.²⁰₋₀.¹⁸, with ~11% of GCs hosting an LMXB at the detection limit, and confirms a broken power-law LF, suggesting no universal LMXB-LF form across early-type galaxies.

ABSTRACT

Chandra X-ray observations routinely resolve tens to hundreds of low-mass X-ray binaries (LMXBs) per galaxy in nearby massive early-type galaxies. These studies have raised important issues regarding the behavior of this population of remnants of the once massive stars in early-type galaxies, namely the connection between LMXBs and globular clusters (GCs) and the nature of the LMXB luminosity function (LF). In this paper, we combine five epochs of Chandra observations and one central field Hubble Space Telescope Advance Camera for Surveys observation of NGC 4697, one of the nearest, optically luminous elliptical (E6) galaxies, to probe the GC-LMXB connection and LMXB-LF down to a detection/completeness limit of (0.6/1.4) x 10^{37} ergs/s. We detect 158 sources, present their luminosities and hardness ratios, and associate 34 LMXBs with GCs. We confirm that GCs with higher encounter rates (Γ_h) and redder colors (higher metallicity Z) are more likely to contain GCs, and find that the expected number of LMXBs per GC is proportional to Γ_h^{0.79+0.18/-0.15} Z^{0.50+0.20/-0.18}, consistent with fainter X-ray sources in Galactic GCs and LMXBs in Virgo early-type galaxies. Approximately 11+/-2% / 8 +/-2% of GCs in NGC 4697 contain an LMXB at the detection/completeness limit. We propose that the larger proportion of metal-rich GCs in NGC 4697 compared to the Milky Way explains why these fractions are much higher than those of the Milky Way at similar luminosities. We confirm that a broken power-law is the best fit to the LMXB-LF, although we cannot rule out a cutoff power-law, and argue that this raises the possibility that there is no universal form for the LMXB-LF in early-type galaxies. We find marginal evidence for different LFs of LMXBs in GCs and the field and different spectra of GC-LMXBs and Field-LMXBs.

Motivation & Objective

• To investigate the relationship between low-mass X-ray binaries (LMXBs) and globular clusters (GCs) in early-type galaxies, particularly in NGC 4697.
• To determine the dependence of LMXB occurrence on GC properties such as encounter rate (Γₕ) and metallicity (Z).
• To characterize the X-ray luminosity function (LF) of LMXBs in NGC 4697 down to a sensitivity limit of 0.6–1.4×10³⁷ erg s⁻¹.
• To compare spectral and luminosity properties of LMXBs in GCs (GC-LMXBs) versus the field (Field-LMXBs).
• To assess whether the LMXB-LF is universal across early-type galaxies or varies by galaxy properties.

Proposed method

• Five epochs of Chandra X-ray observations were combined to achieve deep exposure and improve source detection sensitivity down to 0.6–1.4×10³⁷ erg s⁻¹.
• Hubble Space Telescope Advanced Camera for Surveys (ACS) imaging was used to identify globular cluster (GC) positions and associate X-ray sources with GCs.
• The LMXB luminosity function (LF) was modeled using a broken power-law form, with a break luminosity L_b and two power-law slopes (α_l and α_h) for faint and bright regimes.
• Spectral fitting was performed using X-ray energy bands to derive hardness ratios and model emission with bremsstrahlung and Galactic absorption (N_H), with column density and temperature as key spectral parameters.
• Statistical analysis compared LMXB fractions in GCs and the field, and tested for differences in LFs and spectral properties between GC-LMXBs and Field-LMXBs.
• The dependence of LMXB occurrence on Γₕ and Z was quantified via a power-law fit: expected LMXBs per GC ∝ Γₕ⁰.⁷⁹⁺⁰.¹⁸₋₀.¹⁵ × (Z/Z☉)⁰.⁵⁰⁺⁰.²⁰₋₀.¹⁸.

Experimental results

Research questions

• RQ1 What is the dependence of LMXB occurrence in globular clusters on their encounter rate (Γₕ) and metallicity (Z)?
• RQ2 Is the LMXB luminosity function (LF) in NGC 4697 best described by a broken power-law or a cutoff power-law?
• RQ3 Are there significant spectral differences between LMXBs in globular clusters (GC-LMXBs) and those in the galactic field (Field-LMXBs)?
• RQ4 Do LMXBs above the Eddington limit for a 1.4 M☉ neutron star preferentially reside in GCs?
• RQ5 Is the LMXB-LF in NGC 4697 consistent with a universal form across early-type galaxies?

Key findings

• The expected number of LMXBs per globular cluster in NGC 4697 scales with Γₕ⁰.⁷⁹⁺⁰.¹⁸₋₀.¹⁵ and (Z/Z☉)⁰.⁵⁰⁺⁰.²⁰₋₀.¹⁸, indicating a strong dependence on dynamical interactions and metallicity.
• Approximately 11% ± 2% of globular clusters in NGC 4697 host an LMXB at the detection limit, significantly higher than in the Milky Way due to its higher fraction of metal-rich GCs.
• The LMXB luminosity function is best fit by a broken power-law model, with a break luminosity of (10.8 ± 2.9)×10³⁷ erg s⁻¹, and slopes α_l = 0.83 ± 0.52 (faint end) and α_h = 2.38 ± 0.33 (bright end).
• Spectral analysis shows that GC-LMXBs have higher temperatures and greater absorbing columns than Field-LMXBs, with a significance of 3.3σ, suggesting different accretion or environmental conditions.
• Fainter LMXBs tend to have lower absorbing columns, while brighter LMXBs show a small excess in absorption, possibly due to higher accretion rates or more circumstellar material.
• The spectral model underpredicts soft emission, as indicated by sub-Galactic N_H values in sources with low absorption, suggesting limitations in calibration or modeling at low energies.