Benchmarking transition energies and emission strengths for X-ray astrophysics with measurements at the Livermore EBITs
K-shell transitions in astrophysically abundant metals and L-shell transitions in Fe group elements show characteristic signatures in the soft X-ray spectrum in the energy range 0.1-10 keV. These signatures have great diagnostic value for plasma parameters such as electron and ion temperatures and densities, and can thus help understand the physics controlling the energetic processes in astrophysical sources. This diagnostic power increases with advances in spectral resolution and effective area of the employed X-ray observatories. However, to make optimal use of the diagnostic potential - whether through global spectral modeling or through diagnostics from local modeling of individual lines - the underlying atomic physics has to be complete and well known. With the next generation of soft X-ray observatories featuring micro-calorimeters such as the SXS on Astro-H /Hitomi and the X-IFU on Athena, broadband high-resolution spectroscopy with large effective area will become more commonly available in the next decade. With these spectrometers, the accuracy of the plasma parameters derived from spectral modeling will be limited by the uncertainty of the reference atomic data rather than by instrumental factors, as is sometimes already the case for the high-resolution grating observations with Chandra-HETG and XMM-Newton-RGS. To take full advantage of the measured spectra, assessment of the accuracy of and improvements to the available atomic reference data are therefore important.Dedicated measurements in the laboratory are essential to benchmark the theoretical calculations providing the bulk of the reference data used in astrophysics. Experiments at the Lawrence Livermore National Laboratory electron beam ion traps (EBIT-I and SuperEBIT) have a long history of providing this service. In this work, I present new measurements of transition energies and absolute electron impact excitation cross sections geared towards currently open atomic physics data needs. First, I measured the energies of K α transitions in L-shell ions of Si and S at EBIT using the EBIT calorimeter spectrometer (ECS) with 4.5-5.0 eV resolution, i.e., a similar resolution to the Astro-H /Hitomi SXS soft X-ray spectrometer. While these lines will become interesting also for L-shell ions of other astrophysically abundant elements, they have been observed most prominently from L-shell ions of Si and S in the X-ray spectra of a variety of astrophysical sources. The measured line centers have an accuracy of 0.5 eV for the strong transitions and 1 eV for the weaker ones. This accuracy translates to Doppler shifts of less than 90 km s -1 , i.e., less than the calibration uncertainty of the Chandra high-energy transmission gratings. The measured line centers are identified with my own calculations with the Flexible Atomic Code (FAC) and compared to these and calculations by Palmeri et al. (2008). I demonstrate the impact of these measurements by re-evaluating Doppler shifts for the high-mass X-ray binaries Vela X-1 and Cyg X-1 with the new reference data. Using the high-resolution, imaging focusing spherical crystal spectrometer EBHiX with a quartz 101 crystal, I verified the results from the ECS measurements on K α transitions in N- through Li-like S. The measurement has a spectral resolution of better than 0.52 eV. The derived transition energies have an accuracy of 0.2 eV, corresponding to Doppler shifts of < 30 km s -1 , i.e., within the requirements set by the planned Athena X-ray observatory. Secondly, I used the EBHiX crystal spectrometer with a quartz 110 crystal in second order to measure the strongly blended K α spectra of M-shell Fe ions around 6.4 keV. Contributions to the 6.4 keV line complex from these ions are important for transient plasmas such as those in supernova remnants. While a simple FAC model of Cl- through F-like Fe suggested it should be possible to resolve major contributions from different charge states to this complex at a 2 eV resolution, the data did not allow us to identify any new lines, although the spectral resolution of this measurement was higher than in previous experiments. Thirdly, I measured absolute electron impact excitation (EIE) cross sections for He-like Fe line w and H-like Fe Ly α 1 and Ly α 2 at different electron energies and charge balances. The cross sections of the direct excitation lines are brought to an absolute scale by normalizing to the radiative recombination (RR) spectrum of the same ion. The direct excitation spectrum was measured with the low-energy pixels of the ECS with a spectral resolution of about 6 eV, while the RR spectrum was recorded with the ECS's thicker high-energy pixels with instrumental resolution of about 30 eV. The high-energy pixels thus allowed us, for the first time, to resolve the RR into the n = 2 shell of L-shell Fe ions at electron impact energies high enough to excite a K-shell electron in Fe ions. These measurements of absolute EIE cross sections using the ECS microcalorimeter at EBIT have accuracies on the 10% level, and therefore fulfill the requirements on atomic reference data identified by the astrophysics community. Benchmarking theoretical cross sections on this level tightens the constraints on important diagnostics for, e.g., elemental abundance measurements and resonance scattering in the high-resolution X-ray spectra of the Perseus galaxy cluster observed with Hitomi -SXS. Finally, the performance of the EBHiX crystal spectrometer at EBIT was evaluated for various quartz crystals and the ion temperatures of the ions trapped in EBIT were derived from thermal line broadening measured with EBHiX. The EBHiX's apability to measure the degree of linear polarization for X-ray transitions excited in EBIT was demonstrated for the H-like Mn Ly α line.
- Pub Date:
- March 2017