[Paper Review] Broad-band BeppoSAX observation of the low-mass X-ray binary X1822-371

A. N. Parmar, T. Oosterbroek+5 more|arXiv (Cornell University)|Jan 24, 2000

Astrophysical Phenomena and Observations3 citations

TL;DR

This study presents a broad-band BeppoSAX observation of the low-mass X-ray binary X 1822-371, revealing an unusually complex 0.3–40 keV spectrum best fit by a combination of Comptonized emission and a strong blackbody component contributing over 40% of the 1–10 keV flux. A prominent 1.33 keV absorption edge with optical depth 0.28 is detected, suggesting highly ionized neon or neutral magnesium, though no strong Fe-K or O-K edges are observed, challenging standard accretion disk corona models.

ABSTRACT

Results of a 1997 September 9-10 BeppoSAX observation of the 5.57 hr low-mass X-ray binary (LMXRB) X1822-371 are presented. The 0.3-40 keV spectrum is unusually complex and cannot be fit by any of the standard models applied to other LMXRB. At least two components are required. One component has a shape consistent with that expected from the Comptonization of an input soft (Wein) spectrum while the other, contributing ~40% of the 1-10 keV flux, is consistent with being a blackbody. In addition, there is a ``dip'' in the spectrum which can be modeled by a 1.33 +0.05 -0.11 keV absorption edge with an optical depth, tau, of 0.28 +/- 0.06. If the same model is fit to ASCA Solid-State Imaging Spectrometer spectra obtained in 1993 and 1996, then reasonable fits are also obtained, with a similar absorption feature required. The nature of this feature is highly uncertain; its energy corresponds to the K-edges of highly ionized Ne x and neutral Mg, or to an L-edge of moderately ionized Fe. Surprisingly, no strong (tau > 0.05) Fe-K or (tau > 0.18) O-K edges are visible. The folded lightcurve of X1822-371 is similar to previous observations, except that no strong softening is seen near the eclipse. An updated orbital ephemeris is provided.

Motivation & Objective

• To characterize the broadband X-ray spectrum of the low-mass X-ray binary X 1822-371 using simultaneous BeppoSAX data across 0.3–40 keV.
• To resolve discrepancies in prior spectral fits from HEAO-1, EXOSAT, Ginga, and ASCA, which showed inconsistent features like a dip at ~1.5 keV and anomalous Fe-K line ratios.
• To determine whether the observed spectral complexity arises from intrinsic emission processes or reprocessing in an accretion disk corona (ADC), particularly given the system's edge-on orientation.
• To test if the spectral features—especially the 1.33 keV absorption edge—are stable over time by comparing with earlier ASCA observations.

Proposed method

• Analysis of simultaneous 0.3–40 keV X-ray data from the BeppoSAX satellite, covering the full energy band with high sensitivity.
• Spectral fitting using a combination of a Comptonization model (comptt) and a blackbody component to model the non-thermal and thermal emission, respectively.
• Incorporation of an absorption edge model at ~1.33 keV with variable optical depth to account for the observed spectral dip.
• Comparison of BeppoSAX results with archival ASCA SIS spectra from 1993 and 1996 to assess spectral stability and consistency of the absorption feature.
• Use of hardness ratios and folded lightcurves to assess spectral variability and confirm the absence of strong softening near eclipse.
• Application of the y-parameter formalism to evaluate Comptonization efficiency, with y ≈ 25–40 indicating saturated, Wien-like spectra.

Experimental results

Research questions

• RQ1 What causes the unusually complex X-ray spectrum of X 1822-371, which cannot be fit by standard LMXRB models?
• RQ2 Why does the 1–10 keV flux contain over 40% contribution from a blackbody component, contrary to typical LMXRB behavior?
• RQ3 What is the physical origin of the 1.33 keV absorption edge with optical depth 0.28, and why are strong Fe-K or O-K edges absent?
• RQ4 Is the spectral feature stable over time, as indicated by comparison with earlier ASCA observations?
• RQ5 How do the derived Comptonization parameters (kTe ≈ 5–10 keV, τp ≈ 22–26) compare with correlations observed in globular cluster LMXRBs?

Key findings

• The 0.3–40 keV spectrum of X 1822-371 requires at least two components: a Comptonized continuum and a strong blackbody, with the latter contributing more than 40% of the 1–10 keV flux.
• A distinct absorption edge at 1.33 ± 0.05/0.11 keV with optical depth 0.28 ± 0.06 is required to model a spectral dip, suggesting possible K-edges of highly ionized Ne x or neutral Mg.
• The same absorption feature is required in both 1993 and 1996 ASCA SIS spectra, indicating spectral stability over time, though its physical origin remains uncertain.
• No strong Fe-K (τ > 0.05) or O-K (τ > 0.18) edges are detected, contradicting expectations from photo-ionized ADC models.
• The Comptonization parameters yield a y-parameter of 25–40, indicating a saturated, Wien-like spectrum, consistent with high optical depth and electron temperature of 5–10 keV.
• The derived τp and kTe values are consistent with luminosity-dependent trends observed in globular cluster LMXRBs, suggesting a link between accretion rate and Comptonizing plasma properties.