The Soft X-ray Absorption Excess

The Ionisation State of Gas in High-Redshift Galaxies
The incredibly bright and broadband GRB afterglow illuminates the gas and dust within the star forming regions of the host galaxy, as well as the interstellar material in the disk and halo of the galaxy, and intervening intergalactic medium along the GRB line-of-sight. From the increasing sample of GRB spectroscopic observations that cover the neutral hydrogen Lyman-α absorption feature at UV wavelengths (rest-frame 1215Å), it is becoming clear that GRB host galaxies have high column densities of cold neutral gas (T≤ 103K). Any corresponding ionised hydrogen would not be detected, and neutral hydrogen measurements thus require an ionisation correction to determine the total column density of atomic hydrogen. A large fraction of GRB hosts have such large column densities of neutral hydrogen however, that the ionisation correction is negligible, and the neutral absorbing gas component is known as a damped-Lyman-α (DLA) system (log NHI >20.3cm-2). The survival of certain species, such as Mg I, and time varying Fe II and Ni II fine-structure lines, place this neutral gas component at a few hundred parsecs from the GRB, within the ISM of the host galaxy. The neutral ISM is also traced by low ionisation species detected in the UV, whereas highly-ionised species (e.g. O IV, C IV, Si IV and N V) possibly probe the hot gas (T~104K) within the circumburst environment of the GRB, as well as a contribution from gas in the rest of the galaxy.

In contrast to the specific regions of gas that can be identified from UV spectra, the spectral resolution typically available in X-ray spectroscopic observations limits the information that can be obtained on the location or ionisation state of the absorbing material. However, in contrast to optical and UV wavelengths, X-ray absorption is sensitive to gas at almost all levels of ionisation. The fairly weak dependence between the absorption of X-ray photons by oxygen (the dominant source of X-ray absorption) and the ionisation state of the oxygen makes the absorption measured in the X-ray afterglow a good proxy for the total oxygen column density along the observed line-of-sight.

The regular X-ray afterglow observations available with Swift/XRT over the last 6.5 years have supplied a sample of over 600 GRBs with high signal-to-noise X-ray afterglow spectral observations. These data have revealed a considerable discrepancy between the amount of gas along the line-of-sight inferred from X-ray absorption, and the amount inferred from optical (rest-frame UV) absorption. The former is typically an order of magnitude higher than what is expected given the amount of absorption from neutral element species measured in optical absorption line spectroscopy, suggesting that the X-ray afterglow is absorbed by an additional and significant component of ionised gas that does not affect the UV and optical afterglow.

Figure 1: Left: Host galaxy neutral gas column density, Nntr, against host galaxy total gas column density, NN.O, along the line-of-sight to a sample of 26 GRBs. For each GRB, Nntr is derived from either the column density of Zn II (red circles), S II (orange squares), Si II (green stars), or Fe II (blue triangles), where a correction for dust depletion is applied to NN,Si and NN,Fe. Smaller data points correspond to those take from low- or mid-resolution spectra, and larger data points are taken from high-resolution spectra (R>10,000). The dashed line corresponds wo where Nntr is equal to NN,O. Right: Logarithmic host galaxy column density of the highly ionised atoms C IV, Si IV, N V and O VI against the total logarithmic column densities of C, Si, N and O in the top, second, third and bottom panels, respectively. The total column densities are derived from X-ray observations, denoted by N X. The C IV, Si IV, N V and O VI measurements plotted as circles are all taken from Fox et al. (2008), N V data plotted as squares are from Prochaska et al. (2008), and data taken from D'Elia et al. (2010) are plotted as triangles. In all four panels, the dashed line corresponds to where the normalised soft X-ray column desnity is equal to the column density of the corresponding highly ionised atom.

In Schady et al. (2011) we performed a careful study of the X-ray and optical afterglow spectra, as well as the afterglow broadband SED of 29 GRBs, and showed the neutral gas to consist of <10% of the total gas (see Fig. 1, left panel). Furthermore, the amount of gas inferred from optically measured high-ionisation absorption lines with ionisation potentials of up to ~200eV (e.g. Si IV, C IV, N V , O VI), can only account for a further ~10 % of the gas that absorbs the X-ray afterglow (see Fig. 1, right panel). This result suggests that either the X-ray excess comes from ultra-highly-ionised (ionisation potentials >300 eV) gas within the GRB host galaxy (surrounding ionised bubble or in the galaxy halo), or that the excess X-ray absorption stems from material external to the host galaxy (e.g. intervening systems, or within a local 'warm-host' intergalactic medium, or WHIM.