Highly Ionized Potassium Lines in Solar X-Ray Spectra and the Abundance of Potassium

Abstract

The abundance of potassium is derived from X-ray lines observed during flares by the RESIK instrument on the solar mission CORONAS-F between 3.53Å and 3.57Å. The lines include those emitted by He-like K and Li-like K dielectronic satellites, which have been synthesized using the chianti atomic code and newly calculated atomic data. There is good agreement of observed and synthesized spectra, and the theoretical behavior of the spectra with varying temperature estimated from the ratio of the two GOES channels is correctly predicted. The observed fluxes of the He-like K resonance line per unit emission measure gives log A(K) = 5.86 (on a scale log A(H) = 12), with a total range of a factor 2.9.

THE X-RAY LINE FEATURE AT 3.5 KeV IN GALAXY CLUSTER SPECTRA

“…This should be compared with the result from our previous smaller sample of solar flares (Figure 3 of Sylwester et al 2010b), in which we derived Our estimate of the solar flare abundance relies heavily on the atomic data for helium-like potassium in the CHIANTI database, in particular the values of the effective collision strengths which, in the absence of specific calculations available for the release of CHIANTI v. 7, are based on interpolations of helium-like ions of other elements. Since errors in these values folds directly into the abundance determinations, it is advisable to check the accuracy of the atomic data wherever possible.…”

“…Channel 1 (spectral range 3.26-3.69 keV) includes the He-like K (K XVIII) lines mentioned earlier as well H-like S (S XVI) and Ar (Ar XVIII) Lyman lines and a dielectronic satellite line feature at 3.62 keV due to Ar XVI made up of several lines but with one major contributor. Our earlier analysis (Sylwester et al 2010b) was based on 2795 spectra recorded in 20 flares. For this work, many additional flare spectra were added to a scientifically analyzable level (so-called Level 2: see http://www.cbk.pan.…”

“…Other lines are expected in the region of the K XVIII lines, in particular K XVII dielectronic satellites, the fluxes of which were calculated by Sylwester et al (2010b). The most significant include satellites j and k (3.474, 3.477 keV, respectively), which are blended with K XVIII line z, and d13 and d15 (3.503 keV) which are blended with K XVIII line w. At temperatures 1 < keV, the fluxes of j and k are calculated to be comparable to K XVIII line z but decreasing for higher temperatures (the satellite-to-line w ratio has an approximately T 1 -dependence).…”

“…These are considerably better than the ∼100 eV resolution of the MOS and PN CCD detectors on XMM-Newton. Consequently, the individual members of the K XVIII line complex (w, x y + , z lines: notation of Gabriel 1972) at 3.473, 3.493, and 3.511 keV (photon energies based on wavelengths from Kelly 1987) In this paper, we extend the work of Sylwester et al (2010b) by the addition to RESIK data sets of several thousand spectra, with the consequence that our previous K abundance estimate for all RESIK flares is revised upwards. We then examine whether the spectral signatures recorded by Bulbul et al (2014a) and Boyarsky et al (2014a) are compatible with the K XVIII line emission based on these solar flare observations, and whether there is a need for invoking sterile neutrino decay.…”

“…Unfortunately, these authors have neglected the fact that this region has been seen with high spectral resolution in numerous solar flare spectra with temperatures up to 22 MK (1.9 keV) using the RESIK crystal spectrometer (Sylwester et al 2005(Sylwester et al , 2010b and in the spectra of a binary star system with an active corona (σ Gem) using the Chandra HETG spectrometer (Huenemoerder et al 2013a(Huenemoerder et al , 2013b. The RESIK observations have a spectral resolution of 11 eV (11.5 mÅ), and the Chandra HETG spectra a spectral resolution of 12 eV (12 mÅ).…”