Recent advances in the CHIANTI database in the X-ray range Enrico Landi Naval Research Laboratory
Dec 19, 2015
Recent advances in the CHIANTI database in the
X-ray range
Enrico Landi
Naval Research Laboratory
Overview
1. Introduction
2. What is CHIANTI
3. New features of the CHIANTI database
4. Comparison with X-ray observations
5. Conclusions
Introduction
•X-ray spectra are of primary importance for quantitative studies of the physics of many astrophysical objects
•Emission and absorption of line and continuum radiation in the X-rays offer a wide variety of diagnostic tools to determine the physical properties of emitting sources
•For this reason, in the recent past, several instruments have been flown to observe astrophysical sources in the X-ray range:
Chandra ROSATXMM YohkohRHESSI SMMRESIK …and many others
Why a database?• X-ray spectra are composed by line and continuum radiation,
emitted by highly charged ions.
• Observed X-ray spectral lines come from
Rydberg series in the H-,He-like sequencesHigh-energy configurations (n>2) in highly
charged Fe and Ni ionsInnershell transitionsDielectronic satellite lines
• Continuum radiation comes from
Free-free radiationFree-bound radiation
• In order to study this radiation and use it for plasma diagnostic purposes, a large amount of atomic data are needed for both line and continuum radiation
• To address this need, several databases have been created in the past:
CHIANTI
MEKAL
APEC/APED
ADAS
Arcetri Spectral Code
Requirements for a database
• In order to be suitable for the analysis of modern high-resolution spectra, atomic data bases need to
- be complete no lines left behind
- be accurate plasma diagnostics not hindered by atomic physics uncertainties
- be easy-to-use
- be transparent - the user can independently check the original data and their accuracy
- no black box
- all data independently refereed in peer
reviewed literature
• Also, atomic data and predicted emissivities from data bases need to be benchmarked against observations
The CHIANTI database
• CHIANTI consists of
A database of atomic data and transition rates
A suite of IDL programs for plasma diagnostics
• CHIANTI is able to calculate
Line emissivities for more than 230 ions
innershell transitions
dielectronic satellite lines
Continuum emissivities for free-free radiation
free-bound radiation
two-photon continuum
Under the optically thin plasma assumption
CHIANTI data are:
• In ASCII format
• Selected from the refereed literature (no unpublished data)
• With references to original literature
CHIANTI is completely transparent to the end user
• FREELY available on the web at
http://wwwsolar.nrl.navy.mil/chianti.html
• Fully documented through user guides
CHIANTI also maintains: a mailing list
email assistance to users at:
• CHIANTI has enjoyed great success in the astrophysical community. CHIANTI data have been
– Included in the software of several satellite borne missions
SOHO/CDS (EUV)SOHO/EIT (EUV)TRACE (UV)Solar-B (EUV,X-rays)STEREO (EUV)RHESSI (X-rays)
– Included in other spectral codes
APEC/APEDPintOfAleArcetri Spectral CodeADAS
State-of-the-art
• Literature data for the X-rays have several limitations:
Missing configurations (n>3 in many Fe ions)
Limited atomic models for n=3 configurations (usually due
to computer memory limitation in the past)
Missing processes:resonances
ionization effects on level populations
recombination effects on level populations
cascades from higher levels
Uncertainties in line identifications
CHIANTI 5.0
• The CHIANTI database has been recently greatly expanded. The main features of the next CHIANTI release (Version 5.0) are:
– New data for high-energy configurations in Fe XVII-XXIII
n=3,4,5,6,7 Fe XVII
n=3,4,5 Fe XVIII-XXIII
– New data for satellite lines
– Ionization and recombination effects in level populations
– Complete re-assessment of energy levels and line identifications
– Other data and new ions for EUV and UV lines
New data for high-energy configurations in Fe XVII-XXIII
• We have made use of the Flexible Atomic Code, by Dr. M.F. Gu, to calculate
Energy levelsRadiative transition ratesCollisional transition rates (including resonances)
for all configurations withn=3,4,5,6,7 Fe XVIIn=3,4,5 Fe XVIII-XXIII
• These data allow to predict lines in the 7-12 Angstrom range
• Few if any data were available in the literature for most of these configurations
• Data will be published separately (Landi & Gu 2005)
New data for satellite lines
• New data have been added to CHIANTI 5.0 for dielectronic satellite lines and innershell transitions, to match observations
Fe XXIII innershell transitions
S XVI, Ca XVIII, Fe XXIV dielectronic satellite lines
• These new lines also provide diagnostic tools for measuring the plasma electron temperature
• These lines allow to study RHESSI spectra in the 6-9 keV energy range
Fe XXV
Fe XXV
Fe XXV
Fe XXIVsatellites Fe XXIV
satellites Fe XXIVsatellites
Ionization and Recombination
• Recently, Behar & Doron (2002) and Gu (2003) demonstrated that ionization and recombination are important contributors to steady-state level population in highly ionized Fe ions
• CHIANTI 5.0 incorporates data and software to take these two processes into account
• All recombination and ionization data have been taken from the Flexible Atomic Code calculations by Gu (2003).
• We make use of the Coronal Model Approximation:
Without Ionization/Recombination:
With Ionization/Recombination:
Where: nq-1, nq, nq+1 ion fractions
CI, REC total ion. and rec. rates
Egi total excitation rate level i
Dig total de-excitation rate level i
• The Coronal Model Approximation is not valid if metastable levels have non-negligible population
• As the electron density increases, the population of metastable levels in highly ionized Fe ions also increases
• The maximum density at which metastable level populations are negligible changes from ion to ion:
Ion Log Ne(max)
Fe XVII anyFe XVIII > 13
Fe XIX 12Fe XX 12Fe XXI 12Fe XXII 13Fe XXIII > 13Fe XXIV any
Fe XVII Fe XVIII
Fe XIX Fe XX
Fe XXI Fe XXII
Fe XXIVFe XXIII
Comparison with X-ray observations
• We have compared the new CHIANTI 5.0 with observations of a moderate solar flare
Instrument SMM/FCS (Bragg crystal spectrometer)
Date of observation August 25, 1980Wavelength ranges 13.1-22.4 A (channel 1)
10.6-14.9 A (channel 2)7.3-10.1 A (channel 3)
Spectral resolution 1-20 mA (depending on the channel)Source M 1.5 flareSpectral scan Duration 17.5 minutesIons observed H-like O,Ne,Mg
He-like O,Ne,Na,Mg,AlFe ions Fe XVII to Fe
XXIIINi ions Ni XIX, Ni XX
Comparison method
• FCS spectra were not observed simultaneously, so the flare plasma was analyzed as a function of time
• The Emission Measure analysis showed that
• The flare plasma was isothermal
• The temperature was decreasing slowly
• The emission measure decreased by a factor 6 during the observation
• This allowed us to use the Emission Measure as a tool to compare CHIANTI 5 emissivities and observed fluxes for each ion, to:
• Assess the quality of CHIANTI 5 data
• Identify blends from other ions and evaluate their contribution to the total intensity (additional check on atomic physics)
• Identify areas where improvements are still needed
• In case of isothermal plasma
• We can define, for all the lines of the same ion, the ratio
• If there are no blends and no atomic physics problems, all the ratios must be the same at all temperatures, within the uncertainties.
Example: Fe XIX
CHIANTI 4.2CHIANTI 5
Time bin 1
Time bin 2
Example: Fe XVII
• Long standing problems:
• Strong 15.01 A line lower than predicted
» Resonant scattering?
» Satellites in 15.01/15.26 intensity ratios?
• Disagreement in 2p-3s/2p-3d ratios» Innershell ionization to 3s?
» Satellite contributions to line ratios?
• Existing atomic data» DW collision rates from many authors
» Only very recently resonance data have been considered
» Doron & Behar (2002) showed that recombination is important for Fe XVII
• We used CHIANTI 5.0 to check the importance of many additional processes in Fe XVII level population:
Process Importance
Cascades ModerateCollisional ionization ModerateRecombination CrucialResonances Crucial
• We have compared the FCS spectrum with CHIANTI 5.0 predictions obtained with and without those processes
CHIANTI 4.2
CHIANTI 5.0
15.01 A
15.01 A
Results and Conclusions
• CHIANTI 5.0 reproduces observed high- and low- resolution X-ray spectra with great accuracy
All relevant configurations in Fe ions are now included
Blending from ions of different species is accounted forMost lines are reproduced within 30%
• CHIANTI 5.0 represents a major advance over previous versions and other databases
• New diagnostic tools are now available to measure the physical properties of the emitting plasmas