DRAFT VERSION FEBRUARY 10, 2022 Preprint typeset using L A T E X style emulateapj v. 01/23/15 REVISITING THE Fe XVII LINE EMISSION PROBLEM: HIGH-RESOLUTION LABORATORY MEASUREMENTS OF THE 3s –2p AND 3d –2p LINE-FORMATION CHANNELS CHINTAN SHAH 1, * ,J OSÉ R. CRESPO LÓPEZ-URRUTIA 1 ,MING FENG GU 2 ,THOMAS PFEIFER 1 ,J OSÉ MARQUES 3, 4 ,FILIPE GRILO 4 , J OSÉ PAULO SANTOS 4 , AND PEDRO AMARO 4, ⋆ 1 Max-Planck-Institut für Kernphysik, D-69117 Heidelberg, Germany 2 Space Science Laboratory, University of California, Berkeley, CA 94720, USA 3 University of Lisboa, Faculty of Sciences, BioISI - Biosystems & Integrative Sciences Institute, Lisboa, Portugal and 4 Laboratório de Instrumentação, Engenharia Biomédica e Física da Radiação (LIBPhys-UNL), Departamento de Física, Faculdade de Ciências e Tecnologia, FCT, Universidade Nova de Lisboa, 2829-516 Caparica, Portugal Draft version February 10, 2022 ABSTRACT We determined relative X-ray photon emission cross sections in Fe XVII ions that were mono-energetically excited in an electron beam ion trap. Line formation for the 3s - 2 p and 3d - 2 p transitions of interest proceeds through dielectronic recombination (DR), direct electron-impact excitation (DE), resonant excitation (RE), and radiative cascades. By reducing the electron-energy spread to a sixth of that of previous works and increasing counting statistics by three orders of magnitude, we account for hitherto unresolved contributions from DR and the little-studied RE process to the 3d - 2 p transitions, and also for cascade population of the 3s - 2 p line man- ifold through forbidden states. We found good agreement with state-of-the-art many-body perturbation theory (MBPT) and distorted-wave (DW) method for the 3s - 2 p transition, while in the 3d - 2 p transitions known discrepancies were confirmed. Our results show that DW calculations overestimate the 3d - 2 p line emission due to DE by ∼20%. Inclusion of electron-electron correlation effects through the MBPT method in the DE cross section calculations reduces this disagreement by ∼11%. The remaining ∼9% discrepancy is consis- tent with those found in previous laboratory measurements, solar, and astrophysical observations. Meanwhile, spectral models of opacity, temperature, and turbulence velocity should be adjusted to these experimental cross sections to optimize the accuracy of plasma diagnostics based on these bright soft X-ray lines of Fe XVII. Keywords: atomic data — atomic processes — line: formation — methods: laboratory: atomic — plasmas — X-rays: general 1. INTRODUCTION Soft X-rays spectra from astrophysical hot plasmas at a few MK are recorded by grating spectrometers onboard the Chan- dra and XMM-Newton X-ray observatories. They are dom- inated by the L-shell 3d - 2 p and 3s - 2 p transitions in the 15–17 Å range from Fe XVII (Ne-like ions) (Paerels & Kahn 2003; Canizares et al. 2000) that are used for electron tem- perature, density (Mewe et al. 2001; Behar et al. 2001; Xu et al. 2002; Beiersdorfer et al. 2018), velocity turbulence, and X-ray opacity diagnostics (Brickhouse & Schmelz 2005; Werner et al. 2009; Sanders et al. 2011; de Plaa et al. 2012; Kallman et al. 2014). Decay from the states [2 p 5 1/2 3d 3/2 ] J=1 , [2 p 5 3/2 3d 5/2 ] J=1 , and [2 p 5 3/2 3d 3/2 ] J=1 to the [2 p 6 ] J=0 ground state produces the 3d - 2 p transitions called 3 C,3D, and 3E , respectively. The 3s - 2 p decays known as 3F ,3G, and M2 lines proceed from [2 p 5 1/2 3s 1/2 ] J=1 , [2 p 5 3/2 3s 1/2 ] J=1 , and [2 p 5 1/2 3s 1/2 ] J=2 , also to [2 p 6 ] J=0 . Since the optically thick line 3C and the intercombination line 3D both have low contributions from cascades (Mewe et al. 2001), their intensity ratio mainly depends on direct electron impact excitation (DE) and dielectronic recombina- tion (DR) processes. The diagnostic utility of this ratio is limited by discrepancies between observations (e. g., the re- cent Bernitt et al. (2012)) and various predictions of their os- cillator and collision strengths, which can lead to overestimat- ing opacity effects (Parkinson 1973; Brown et al. 1998; Lam- * chintan@mpi-hd.mpg.de ⋆ pdamaro@fct.unl.pt ing et al. 2000; Beiersdorfer et al. 2002, 2004; Brown et al. 2006; Chen 2007; Liang & Badnell 2010; Gillaspy et al. 2011; Bernitt et al. 2012). While nonlinear effects in certain scenar- ios could also reduce the 3C/3D ratio (Oreshkina et al. 2014; Loch et al. 2015), and in spite of forty years of efforts, the 3C/3D collision-strength discrepancy remains essentially un- solved (Chen 2011; Santana et al. 2015; Mendoza & Bautista 2017; Wang et al. 2017). Alternative approaches use the forbidden M2 line or the 3G + M2 line blend for opacity, and turbulence-velocity di- agnostics, since the M2 line is optically thinner than the 3D line (de Plaa et al. 2012; Werner et al. 2009). However, astro- physical observations of the (3G + M2)/3C ratio and 3s/3d or (3F + 3G + M2)/(3C + 3D + 3E ) ratio also depart from spectral models (de Plaa et al. 2012), and laboratory ratios are consistently larger than the calculated ones (Beiersdorfer et al. 2002, 2004). This could be explained if M2 is mainly fed by complex cascades following DE, with contributions from resonant excitation (RE) (Beiersdorfer et al. 1990; Doron & Behar 2002; Gu 2003; Tsuda et al. 2017). Another argu- ment points to the same cause than the 3C/3D discrepancy. Thus, laboratory validation of calculations of the DR, DE, and RE contributions to these transitions is essential to con- struct reliable spectral models, and urgently needed in view of the upcoming high-resolution space missions XRISM Re- solve (Tashiro et al. 2018), Arcus (Smith et al. 2016), and Athena (Barret et al. 2016). In this Letter, we perform calculations both distorted-wave (DW) and many-body perturbation theory (MBPT) of the Fe XVII 3s (3F + 3G + M2) and 3d (3C + 3D + 3E ) emissions arXiv:1903.04506v1 [astro-ph.HE] 11 Mar 2019