2004ApJ...605..793F

We searched for absorption lines of highly ionized O and Ne in the energy spectra of two low-mass X-ray binaries, 4U 1820-303 in the globular cluster NGC 6624 and Cyg X-2, observed with the Chandra LETG, and detected O VII, O VIII, and Ne IX absorption lines for 4U 1820-303. The equivalent width of the O VII Kα line was 1.19^+0.47_–0.30eV (90% errors), and the significance was 6.5 σ. Absorption lines were not detected for Cyg X-2 with a 90% upper limit on the equivalent width of 1.06 eV for O VII Kα. The intrinsic line width was not resolved, and an upper limit corresponding to a velocity dispersion of b=420km/s was obtained for the O VII Kα line of 4U 1820-303. The ion column densities were estimated from the curve-of-growth analysis, assuming several different values of b. The absorption lines observed in 4U 1820-303 are likely due to hot interstellar medium, because O will be fully photoionized if the absorbing column is located close to the binary system. The velocity dispersion is restricted to b=200-420km/s from consistency between O VII Kα and Kβ lines, the Ne/O abundance ratio, and H column density. The average temperature and the O VII density are estimated to be logT(K)=6.2-6.3 and n_OVII_=(0.7-2.3)x10^–6cm^–3, respectively. The difference of O VII column densities for the two sources may be connected to the enhancement of the soft X-ray background (SXB) toward the Galactic bulge region. Using the polytrope model of hot gas to account for the SXB, we corrected for the density gradient and estimated the midplane O VII density at the solar neighborhood. The scale height of hot gas is then estimated using the active galactic nuclei (AGN) absorption lines. It is suggested that a significant portion of both the AGN absorption lines and the high-latitude SXB emission lines can be explained by the hot gas in our Galaxy.