High Spectral Resolution X-ray Observations of the Evolved Supermassive Stellar Binary System $η$ Carinae - Iron K$α$ Band Profile Revealed with XRISM

1. Problem and Scientific Context

The massive binary system $\eta$ Carinae consists of a primary Luminous Blue Variable (LBV) star ($\sim90 M_\odot$) and a massive companion (likely an O-supergiant or Wolf-Rayet star). The system drives the most vigorous wind-wind collision (WWC) in the solar neighborhood, creating shock-heated plasma that emits X-rays.

• The Challenge: Previous high-resolution X-ray observations (e.g., Chandra/HETGS) lacked the energy resolution ($\sim30$ eV at 6 keV) and collecting area to fully resolve the complex Fe K$\alpha$ band (6–7 keV). Consequently, the distinct velocity components of the thermal plasma, the nature of the fluorescent lines, and the presence of Compton scattering features remained poorly constrained.
• The Goal: To utilize the unprecedented energy resolution of the XRISM/Resolve microcalorimeter to dissect the Fe K$\alpha$ band profile, determine the kinematics of the post-shock plasma, probe the geometry of the WWC, and constrain the elemental abundances (specifically hydrogen depletion) of the primary wind.

2. Methodology

• Observations: The XRISM observatory observed $\eta$ Carinae twice:
  • XRI231123: November 23, 2023 (Orbital phase $\phi_X = 0.68$) during the commissioning phase.
  • XRI240609: June 9, 2024 (Orbital phase $\phi_X = 0.78$) during the performance verification phase.
  • Note: The XRI231123 observation suffered from a gate valve issue, limiting sensitivity to $>1.7$ keV and requiring special gain recalibration using a calibration pixel. XRI240609 provided superior statistics and gain accuracy ($\lesssim0.1$ eV).
• Data Analysis:
  • Spectra were extracted from the Resolve instrument (Hp-grade events) between 1.7–12 keV, with a specific focus on the 6.2–7.1 keV Fe K$\alpha$ band.
  • Non-X-ray background (NXB) was modeled and subtracted.
  • Spectral Modeling: A multi-component model was developed in XSPEC:
    • Thermal Components: Three distinct plasma components (Narrow, Blue Wing, Red Wing) modeled with bapec (collisional ionization equilibrium) and convolved with a custom empirical asymmetric smoothing function (htsmooth) to reproduce the non-Gaussian line wings.
    • Fluorescent Lines: Modeled using Lorentzian profiles (feklor/Holzer et al. 1997) for Fe K$\alpha$ and K$\beta$, smoothed by gsmooth.
    • Compton Scattering: A custom empirical model (emcomp) was developed to simulate the down-scattering (Compton shoulder) of both thermal and fluorescent lines, accounting for backscattering geometry.
  • Statistical Analysis: Fits were performed using C-statistics with Markov Chain Monte Carlo (MCMC) to derive 90% confidence intervals.

3. Key Contributions

• First Clear Detection of Thermal Compton Shoulders: The paper reports the first unambiguous detection of Compton shoulders associated with thermal emission lines (Fe He$\alpha$ and Ly$\alpha$) from a celestial source, distinct from the more commonly seen fluorescent line shoulders.
• Resolution of Velocity Components: For the first time, the Fe K$\alpha$ complex is resolved into three distinct kinematic components: a narrow core and two high-velocity wings (blue and red).
• Geometric Constraints: The analysis provides robust evidence that the companion star is on the observer's side (near side) during these observations, while the primary wind is on the far side.
• Abundance Measurement: A novel method using the ratio of fluorescent line flux to Compton-scattered continuum is used to derive the electron-to-iron ratio, revealing severe hydrogen depletion in the primary wind.

4. Key Results

A. Thermal Plasma Kinematics (Fe He$\alpha$ and Ly$\alpha$)

The spectra reveal three velocity components for the $\sim4$ keV thermal plasma:

1. Narrow Component: Centered near zero radial velocity ($\sim -13$ km s$^{-1}$) with a Gaussian broadening of $\sigma \approx 290$ km s$^{-1}$. This likely originates from the stagnation point of the WWC or plasma penetrating the primary wind.
2. Blue Wing: Extends to a maximum blueshift of $\sim -2220$ km s$^{-1}$.
3. Red Wing: Extends to a maximum redshift of $\sim +3100$ km s$^{-1}$.

• Interpretation: The high-velocity wings correspond to the post-shock companion wind flowing along the contact discontinuity. The triangular line profile (rather than a double-horn) suggests significant turbulence or a shock cone opening angle near $180^\circ$.

B. Fluorescent Lines and Compton Scattering

• Fluorescence: The Fe K$\alpha$ (6.4 keV) and K$\beta$ (7.06 keV) lines are significantly blueshifted ($\sim -126$ km s$^{-1}$) and broadened ($\sim 181$ km s$^{-1}$). This confirms the fluorescing material is the primary wind moving toward the observer.
• Compton Shoulders:
  • A clear Compton shoulder is detected for the Fe He$\alpha$ line (peaking at $\sim 6.54$ keV) and marginally for the K$\alpha$ line.
  • The scattering profile indicates strong backscattering and minimal forward scattering, confirming the scattering medium (primary wind) is located behind the emitting source (WWC plasma) relative to the observer.

C. Elemental Abundance and Hydrogen Depletion

• Method: The Equivalent Width (EW) of the Fe K$\alpha$ fluorescence line relative to the Compton-scattered continuum was measured.
• Result: The observed EW ($\sim 947$ eV) is significantly higher than the theoretical value for solar abundances ($\sim 580$ eV).
• Implication: Since the EW is proportional to the number of Fe atoms per electron, the high EW implies a severe depletion of electrons. This is best explained by the CNO cycle converting Hydrogen to Helium (4 H $\to$ 1 He), reducing the electron count.
• Quantification: The data suggests a Hydrogen abundance of $\sim 0.16$ solar (H mass ratio $\sim 0.12$) and a Nitrogen-to-Helium ratio consistent with CNO-processed material. This indicates the primary star is in a late LBV or Nitrogen-rich Wolf-Rayet phase, potentially starting core Helium burning.

5. Significance

• Astrophysical Shock Physics: The study provides the most detailed view of wind-wind collision dynamics to date, confirming that the hottest X-ray plasma originates from the companion's fast wind and revealing the complex flow geometry (turbulence vs. laminar flow).
• Stellar Evolution: The detection of CNO-processed material in the primary wind of $\eta$ Carinae offers direct evidence of advanced nuclear processing and mass loss history, refining models of massive star evolution and LBV behavior.
• Instrumental Milestone: This work demonstrates the transformative power of XRISM/Resolve in resolving complex line profiles and detecting subtle scattering features (Compton shoulders) that were previously invisible, setting a new standard for X-ray spectroscopy of massive binaries.
• Geometry Resolution: By combining line shifts and scattering profiles, the paper definitively resolves the viewing geometry of the binary system, correcting previous ambiguities regarding the orientation of the WWC cone.