Previous
abstract 
Next abstract
We obtained far-ultraviolet spectra of the quasar E1821+64 using the Hopkins Ultraviolet Telescope during the flight of Astro-1 aboard the space Shuttle Columbia in December 1990. The HUT spectra cover the 912-1860 \AA\ wavelength range with a resolution of $\sim$3 \AA. In two successive orbits we accumulated 2966 s of high quality data during orbital night. Our spectra extend well past the peak of the blue bump down to a wavelength of 705 \AA\ in the rest frame of this z=0.297 quasar. There is no significant feature or change in spectral shape in the vicinity of the Lyman edge. We model the spectrum with a relativistic accretion disk with black body emissivities computed in the Schwarzschild metric. Our best fit gives a central black hole mass of $\sim 13 \times 10^{8}~M_\odot$ accreting mass at a rate of 19 $M_\odot~yr^{-1}$. This accretion disk spectrum cannot account for the soft X-ray excess observed with ROSAT reported by Kolman et al., nor can Comptonized disk spectra simultaneously match the definite drop in UV flux and the soft X-ray excess.
The strength of the soft X-ray excess depends sensitively on the hydrogen column density along the line of sight. From archival HST/FOS spectra we measure a column density using the observed galactic damped Ly$\alpha$ profile of $1.8 \times 10^{20}~cm^{-2}$. This is significantly less than the $4.1 \times 10^{20}~cm^{-2}$ suggested by the Stark et al. survey. For a column density as low as $1.9 \times 10^{20}~cm^{-2}$, Kolman et al. find that all the soft X-ray flux seen with ROSAT can be attributed to the central star of the nearby planetary nebula K1-16. We suggest that there is no significant soft X-ray excess in E1821+64, consistent with the relatively weak C\ IV and O\ VI emission in our spectra compared to Ly$\alpha$.
