E. Träbert (*#), P. Beiersdorfer(*), J. Clementson (*)
(*)Lawrence Livermore National Laboratory, Livermore CA, USA. (#)Astronomisches Institut, Ruhr-Universität Bochum, Germany
A so-called XUV excess has been suspected with the relative fluxes of Fe XVIII and Fe XIX lines observed in the XUV and EUV ranges of the spectrum of the star Capella as observed by the Chandra spacecraft, even after correction for interstellar absorption . This excess becomes apparent in the comparison of the observations with simulations of stellar spectra obtained using collisional-radiative models that employ, for example, the Atomic Plasma Emission Code (APEC) or the Flexible Atomic Code (FAC). We have addressed this problem by laboratory studies using the Livermore electron beam ion trap (EBIT), trying to find out whether the purported problem might be caused by insufficient knowledge of the atomic data or of the implementation of atomic processes in plasma modeling. Our understanding of the EBIT spectrum is founded on detailed Fe spectra that have mostly be obtained by Brown et al.  at the same EBIT. The electron density of the exciting electron beam in an EBIT can be adjusted in the range of about 1011 to 1013 cm-3. This is lower than in a typical tokamak plasma and compatible to the density in energetic solar flares. The quiet solar corona has a somewhat lower density again. Apparently EBIT is closest of all laboratory plasmas to the conditions in many astrophysical plasmas.
In our present experiments, the relative detection efficiency between the EUV (near 100 Å) and XUV (near 16 Å) ranges has been determined on the basis of the calculated branching ratio of 1-3 and 2-3 transition in the H-like spectrum O VIII. Inside the EUV range, the wavelength dependence of the grating efficiency of a flat-field diffraction grating similar to ours has previously been measured at a synchrotron light source. In the XUV, the wavelength dependence of the detection efficiency relates to calculated line intensities along the 1s-np Rydberg series of O VIII. FAC calculations assuming several electron beam energies, electron densities, and excitation by a monoenergetic electron beam or a Maxwellian electron energy distribution serve to investigate the theoretical line intensity pattern as well as providing a correction of the observations made with an electron beam (in EBIT) for the Maxwellian excitation in a natural plasma. In the EUV, the line intensity pattern predicted by FAC for EBIT conditions agrees reasonably well with the laboratory and Capella observations. In the XUV wavelength range, agreement of laboratory and astrophysical line intensities is patchy. The spectral simulation results from FAC are much closer to observation than those obtained by APEC. Overall, instead of claiming an XUV excess the XUV/EUV line intensities can be interpreted as suggesting a somewhat higher temperature of Capella than the previously assumed T = 6 MK.
This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344 and was supported by the NASA under work order NNH07AF81I issued by the APRA Program. E.T. acknowledges support by Deutsche Forschungsgemeinschaft, Germany.
Paper presented at the 11th International Conference on Atomic Processes in Plasmas, Queen’s University Belfast, 19-22 July 2011.