The Galactic dust-to-metals ratio and metallicity using gamma-ray bursts

The metallicity and dust-to-metals ratio of the Galaxy are fundamental parameters in understanding the ISM, but there is still uncertainty surrounding these parameters. In this paper, the dust-to-metals ratio in the Galaxy is determined using the photoelectric absorption of the X-ray afterglows of a sample of several hundred gamma-ray bursts (GRBs) to determine the metal column density in combination with Galactic dust maps to determine the line-of-sight dust extinction through the Galaxy in the direction of the GRB. GRB afterglows often have large extragalactic soft X-ray absorptions and therefore the GRB sample’s upper-bound will define the Galactic dust-to-metals relation. Using a two-dimensional two-sample KS test, we determine this upper-bound and so derive the dust-to-metals ratio of the Galaxy. We find N_H = 2.2^{+0.3}_{-0.4}e21 cm^-2 A_V assuming solar, Anders & Grevesse (1989), metallicity. This result is consistent with previous findings using bright X-ray sources in the Galaxy. Using the same technique but substituting the HI maps from the Leiden-Argentine-Bonn survey for the dust maps, allows us to place a limit on the metallicity in the Galaxy. We find a metallicity consistent with the Anders & Grevesse (1989) solar values often used in X-ray fitting. Based on this and previous studies, we suggest that the metallicity of a typical ISM sightline through the Galaxy is ~0.25 dex higher than the current best estimate of the solar metallicity. We further show that the dust-to-gas ratio seems to be correlated with the total gas column density, and that this may be due to the metallicity gradient observed toward the Galactic centre. Based on the non-constant nature of the dust-to-gas ratio, we propose that the dust column density, at N_H = 2.2e21 cm^-2 A_V, represents a better proxy for the soft X-ray absorption column density than HI maps.

💡 Research Summary

The paper presents a novel determination of the Milky Way’s dust‑to‑metals ratio and overall metallicity by exploiting the soft X‑ray absorption measured in the afterglows of a large sample of gamma‑ray bursts (GRBs). Using all Swift‑observed GRBs up to November 2010 (638 events), the authors fit each X‑ray spectrum with a power‑law plus photo‑electric absorption model (phabs*pow) to extract an equivalent hydrogen column density, NH_X, which actually traces the total metal column along the line of sight. Because GRBs are extragalactic and extremely bright, their afterglows sample random sightlines through the Galaxy and provide a lower envelope of absorption that corresponds to the Galactic contribution alone, while the extragalactic component adds a positive offset.

For each GRB direction the authors also retrieve two independent Galactic foreground measurements: (i) the dust extinction AV from the Schlegel, Finkbeiner & Davis (1998) all‑sky dust map (corrected by 14 % following Schlafly & Finkbeiner 2010) and (ii) the atomic hydrogen column N_HI from the Leiden‑Argentine‑Bonn (LAB) 21 cm survey (Kalberla et al. 2005). AV is converted from E(B–V) assuming a standard R_V = 3.1, while N_HI uncertainties are taken as 10 %.

The core statistical technique is a two‑dimensional two‑sample Kolmogorov–Smirnov (2D‑2SKS) test. The authors model the distribution of points in the NH_X–AV and NH_X–N_HI planes as a sum of a Galactic component (log‑normal in AV or N_HI) and an extragalactic component (also log‑normal). By stepping through the parameter space and maximizing the KS probability, they locate the upper envelope that best separates the Galactic contribution from the extragalactic tail. The normalisation of this envelope directly yields the dust‑to‑metals ratio.

Assuming solar abundances from Anders & Grevesse (1989, AG89), the best‑fit relation is

 NH_X = (2.2 +0.3 −0.4) × 10²¹ cm⁻² AV⁻¹

and the ratio of X‑ray to 21 cm columns is

 NH_X / N_HI = 1.10 ± 0.20.

These values agree with earlier studies that used bright X‑ray binaries or supernova remnants (e.g., Predehl & Schmitt 1995). When the authors repeat the analysis with the more recent ISM abundance set of Wilms, Allen & McCray (2000) (implemented via the tbabs model), the derived ratios increase by roughly 40 %, confirming that the dominant systematic uncertainty is the assumed metallicity.

From the NH_X–N_HI comparison the authors infer that the typical Galactic sightline has a metallicity about 0.25 dex higher than the current best estimate of the solar photospheric abundance (Asplund et al. 2009). This suggests that the interstellar medium (ISM) in the solar neighbourhood is modestly metal‑rich relative to the Sun, or that the Sun’s composition is slightly sub‑solar.

An additional finding is that the dust‑to‑gas ratio (NH_X / AV) appears to increase with total column density. The authors propose that this trend reflects the well‑known metallicity gradient toward the Galactic centre: higher metallicity yields more dust per unit gas, thus steepening the NH_X–AV relation at larger NH. Consequently, they argue that dust extinction (AV) is a more reliable proxy for soft X‑ray absorption than the HI column, especially in regions of high column density where the HI map may saturate or miss molecular contributions.

Strengths of the work include (1) a truly all‑sky, unbiased sample of sightlines; (2) the ability to probe very high extinctions (AV ≫ 10 mag) inaccessible to UV/optical studies; and (3) a robust statistical method that isolates the Galactic envelope without needing to model each individual GRB host. Limitations are the reliance on a single set of dust and HI maps (both with known systematic uncertainties), the assumption of neutral gas for the X‑ray absorber, and the neglect of possible low‑level contributions from the intergalactic medium or intervening absorbers—though the authors argue these effects are negligible for the bulk of the sample.

In summary, the paper provides an updated, statistically solid measurement of the Milky Way’s dust‑to‑metals ratio (NH_X/AV ≈ 2.2 × 10²¹ cm⁻² mag⁻¹) and suggests a modestly supersolar ISM metallicity. It also highlights that dust extinction maps may serve as a superior predictor of soft X‑ray absorption compared with traditional 21 cm HI surveys, a conclusion that could improve Galactic absorption corrections in future X‑ray astronomy.