Charge Exchange X-ray Emission of M82: K{alpha} triplets of O VII, Ne IX, and Mg XI

Starburst galaxies are primary feedback sources of mechanical energy and metals, which are generally measured from associated X-ray emission lines assuming that they are from the thermal emission of the outflowing hot gas. Such line emission, however, can also arise from the charge exchange X-ray emission (CXE) between highly ionized ions and neutral species. To understand the feedback of energy and metals, it is crucial to determine the origin of the X-ray emission lines and to distinguish the contributions from the CXE and the thermal emission. The origin of the lines can be diagnosed by the K{\alpha} triplets of He-like ions, because the CXE favors the inter-combination and forbidden lines, while the thermal emission favors the resonance line. We analyze the triplets of O VII, Ne IX, and Mg XI observed in the XMM- Newton reflection grating spectra of the starburst galaxy M82. The flux contribution of the CXE is 90%, 50%, and 30% to the O VII, Ne IX, and Mg XI triplet, respectively. Averaged over all the three triplets, the contribution of the CXE is \sim 50% of the total observed triplet flux. To correctly understand the hot outflow of starburst galaxies, it is necessary to include the CXE. Based on the measured CXE contributions to the O VII, Ne IX, and Mg XI triplets, we estimate the relative abundances of O, Ne, and Mg of the outflow and find they are similar to the solar ratios.

💡 Research Summary

The paper investigates the origin of the prominent He‑like Kα triplet lines (O VII, Ne IX, Mg XI) observed in the X‑ray spectrum of the prototypical starburst galaxy M82, with the aim of quantifying the relative contributions of charge‑exchange X‑ray emission (CXE) and thermal plasma emission. The authors use archival XMM‑Newton Reflection Grating Spectrometer (RGS) observations, combining four pointings (ObsIDs 0112290201, 0206080101, 0560590201, 0560590301) to achieve an effective exposure of roughly 120 ks after filtering out periods of high background. Because the RGS slit is not aligned with the outflow axis, the spatial extent of the source dominates the observed line widths; therefore the authors restrict their analysis to a central 30″ region in the cross‑dispersion direction and also extract O VII triplet spectra from two off‑center regions (−90″ to −30″ and +30″ to +60″) where the line is still detectable.

The diagnostic power of He‑like triplets lies in the different line‑ratio signatures of thermal collisional excitation versus CXE. In a hot, collisionally‑ionized plasma, electron impact preferentially populates the resonance (r) transition, making the r line the strongest. In contrast, CXE—where a highly ionized ion captures an electron from a neutral atom or molecule—populates high‑n levels that cascade preferentially through the inter‑combination (i) and forbidden (f) lines, enhancing these components relative to the resonance line. By fitting the observed triplets with a superposition of a CXE component (based on laboratory measurements of the O VII CXE spectrum from Beiersdorfer et al. 2003) and a thermal component (modeled with APEC at a temperature of 8 × 10⁶ K), the authors can directly infer the flux contributed by each process.

For the O VII triplet, the forbidden line dominates the profile in all three spatial regions, and the best‑fit model attributes 90 % ± 12 % of the total O VII flux to CXE, with only a minor thermal contribution (∼10 %). The consistency of this result across the central and off‑center regions indicates that CXE is occurring not only in the starburst nucleus but also throughout the outflow. The Mg XI triplet shows a different behavior: the resonance line is strongest, suggesting a larger thermal component. Nevertheless, fitting the Mg XI profile yields a CXE contribution of about 30 % ± 15 %. The temperature of the hot gas is independently constrained by the ratio of Mg XII Lyα (5.6 × 10⁻⁵ ph s⁻¹ cm⁻²) to the Mg XI triplet (9.7 × 10⁻⁵ ph s⁻¹ cm⁻²), which corresponds to a plasma temperature of roughly 8 × 10⁶ K. Incorporating the CXE component does not significantly alter this temperature estimate, confirming the robustness of the thermal model.

The Ne IX triplet is more complex because it is blended with several Fe XIX lines. The authors therefore include additional Gaussian components to represent the Fe XIX contribution and explore two plausible Ne/Fe abundance ratios (1.5 × solar, as derived by Ranalli et al. 2008, and 4 × solar, motivated by the unusually high Mg/Fe ratio reported). In both cases the fitted CXE fraction remains near 50 % of the total Ne IX flux, indicating that the exact Ne/Fe ratio does not strongly affect the CXE estimate.

A key part of the analysis is the treatment of higher‑order charge‑exchange cascades. The authors argue that, at the inferred temperature of ∼8 × 10⁶ K, O VIII is scarce while O IX dominates the ion population. Multiple charge‑exchange collisions (O IX → O VIII → O VII) are therefore likely, and the CXE flux in O VIII Lyα should be comparable to that in the O VII triplet. Using the measured O VII CXE flux (8.6 × 10⁻⁵ ph s⁻¹ cm⁻²), they infer that roughly half of the observed O VIII Lyα emission originates from CXE. A similar argument is applied to Mg, where the CXE contribution to Mg XII Lyα is estimated to be about 40 % of the Mg XI triplet flux.

Finally, the authors use the CXE‑derived fluxes of the three triplets to estimate the relative abundances of O, Ne, and Mg in the outflow. After correcting for the CXE contribution, the derived abundance ratios are consistent with solar values (Lodders 2003). This result contrasts with previous studies that assumed purely thermal emission and reported supersolar α‑element to Fe ratios.

In summary, the paper demonstrates that charge‑exchange processes contribute substantially—on average ∼50 %—to the He‑like triplet emission in M82. Ignoring CXE leads to overestimates of metal abundances and potentially mischaracterizes the thermal state of the outflow. The work underscores the necessity of incorporating CXE into spectral models of starburst‑driven galactic winds, both for accurate diagnostics of feedback energetics and for reliable measurements of chemical enrichment in the circumgalactic medium.