X-ray absorption lines of highly-ionized species such as O vii at about zero redshift have been firmly detected in the spectra of several active galactic nuclei .
However , the location of the absorbing gas remains a subject of debate .
To separate the Galactic and extragalactic contributions to the absorption , we have obtained Chandra LETG-HRC and Far Ultraviolet Spectroscopic Explorer observations of the black hole X-ray binary LMC X–3 .
We clearly detect the O vii K \alpha absorption line with an equivalent width of 20 ( 14 , 26 ) mÅ ( 90 % confidence range ) .
The Ne IX K \alpha absorption line is also detected , albeit marginally .
A joint analysis of these lines , together with the non-detection of the O vii K \beta and O viii K \alpha lines , gives the temperature , velocity dispersion , and hot oxygen column density as 1.3 ( 0.7 , 1.8 ) \times 10 ^ { 6 } K , 79 ( 62 , 132 ) { ~ { } km~ { } s ^ { -1 } } , and 1.9 ( 1.2 , 3.2 ) \times 10 ^ { 16 } { ~ { } cm ^ { -2 } } , assuming a collisional ionization equilibrium of the X-ray-absorbing gas and a Galactic interstellar Ne/O number ratio of 0.18 .
The X-ray data allow us to place a 95 % confidence lower limit to the Ne/O ratio as 0.14 , but the upper limit is not meaningfully constrained .
The O vii line centroid and its relative shift from the Galactic O i K \alpha absorption line , detected in the same observations , are inconsistent with the systemic velocity of LMC X–3 ( +310 { ~ { } km~ { } s ^ { -1 } } ) .
The far-UV spectrum shows O vi absorption at Galactic velocities , but no O vi absorption is detected at the LMC velocity at > 3 \sigma significance .
The measured Galactic O vi column density is higher than the value predicted from the O vii-bearing gas , indicating multi-phase absorption .
Both the nonthermal broadening and the decreasing scale height with the increasing ionization state further suggest an origin of the highly-ionized gas in a supernova-driven galactic fountain .
In addition , we estimate the warm and hot electron column densities from our detected O II K \alpha line in the LMC X–3 X-ray spectra and from the dispersion measure of a pulsar in the LMC vicinity .
We then infer the O/H ratio of the gas to be \gtrsim 8 \times 10 ^ { -5 } , consistent with the chemically-enriched galactic fountain scenario .
We conclude that the Galactic hot interstellar medium should in general substantially contribute to zero-redshift X-ray absorption lines in extragalactic sources .