sequences Peter Young George Mason University VA Uri Feldman Artep Inc MD Work funded by NSF and NASA The CHIANTI atomic database Atomic data and software package for modeling ID: 211214
Download Presentation The PPT/PDF document "Assessing atomic data accuracy along iso..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Assessing atomic data accuracy along isoelectronic sequences
Peter Young*George Mason University, VAUri FeldmanArtep Inc, MD
*Work funded by NSF and NASASlide2
The CHIANTI atomic databaseAtomic data and software package for modeling emission linesions and neutrals for elements up to zincsoftware written in IDL and PythonFirst released in 1996 (Dere
et al. 1997, A&AS)Latest version: 7.1 (Landi et al. 2013, ApJ)Citations: 1943Large effort gone into benchmarking atomic data against observationsCHIANTI atomic data are used in other codes (CLOUDY, APED, MOCASSIN, XSTAR)Dr Peter Young (GMU)http://pyoung.org/talks/aas20142
http://
chiantidatabase.org
Team members
Ken
Dere
Giulio
Del
Zanna
Enrico
Landi
Helen Mason
Peter YoungSlide3
The CHIANTI atomic databaseCHIANTI contains atomic data for solving level balance equations for atoms and ionsPrincipal data-sets are:electron excitation rates
radiative decay ratesexperimental energy levelsDr Peter Young (GMU)http://pyoung.org/talks/aas20143Slide4
Level populationsLevel balance equations are αji are atomic rate coefficients (radiative decay rates, electron excitation rates, etc.)n
j are the level populations (normalized so that Σ nj = 1)Dr Peter Young (GMU)http://pyoung.org/talks/aas20144
n
j
computed with CHIANTI IDL routines
pop_solver.pro
&
show_pops.pro
Method is to plot the
n
j
along an isoelectronic sequence
Set temperature to be
T
max
of ion
Use density 10
10
cm
-3Slide5
Li-like ionsExample of a ‘good’ caseDr Peter Young (GMU)http://pyoung.org/talks/aas2014
5Slide6
He-like ionsAn atomic data error clearly seen for N VIDr Peter Young (GMU)http://pyoung.org/talks/aas2014
62-photon transition had been incorrectly assignedFixed with recent v7.1.4 releaseSlide7
C-like ions: the “iron bump”For 2p3 levels an apparent anomaly for iron is seenDr Peter Young (GMU)http://pyoung.org/talks/aas2014
7Slide8
High-Z ions vs. low-Z ionsThe excited ground configuration levels of high-Z ions gain significant population only at high densitiesConsider 2s2p3 3P0 (index=10) level of C-like ions
Dr Peter Young (GMU)http://pyoung.org/talks/aas20148=> High-Z ion models are much more sensitive to the accuracy of weak transitions
Si IX (log T = 6.0)
Fe XXI (log T = 7.1)
Level
n
i
q
i,10
n
i
q
i,10
n
i
q
i,10
n
i
q
i,10
3
P
0
0.10
6.9(-12)
6.9(-13)
1.0
6.9(-13)
6.9(-13)
3
P
1
0.31
5.9(-10)
1.8(-10)
9.2(-4)
8.0(-11)
7.4(-14)
Weak
3
P
0
–
3
P
0
transition is dominant excitation channel
Density=10
10
cm
-3Slide9
Fe XXI 2s 2p3 3P0Why is the population for Fe XXI 2s 2p3 3P0 enhanced?
Fe XXI has 621 levels; Mn XX has 20 levelsDr Peter Young (GMU)http://pyoung.org/talks/aas20149Mn
XX
Fe XXI
Contrib.
Level
Process
Contrib.
Level
Process
56.8%
1
e exc.
49.2%
1
e exc.
42.8%
2
e
exc.
26.0%
25
cascade
4.9%
2
e exc.
3.9%
26
cascade
The ground level excitation rate is much higher for Fe XXISlide10
Collision strength comparisonBig difference in collision strengths for weak ground level excitationDr Peter Young (GMU)http://pyoung.org/talks/aas2014
10Mn XX: Zhang & Sampson (1996, ADNDT, 63, 275) – distorted waveFe XXI: Badnell & Griffin (2001, J.Phys.B, 34, 681) – R-matrix
Iron bump caused by
enhanced direct excitation rates
cascading from high levelsSlide11
ImplementationWith CHIANTI 8 (coming soon!) we will distribute an IDL routine to plot isoelectronic level populationsFor investigating population processes of a specific level, use
Dr Peter Young (GMU)http://pyoung.org/talks/aas201411IDL> ch_plot_iso_pops, ‘o_6’, 3
(plots isoelectronic populations for the level identified as no. 3 for O VI -> 1s
2
2p
2
P
3/2
)
IDL>
pop_processes
, ‘o_6’,
lev
=3, /verbose
Into
level
Out of level
Radiative
decay
0.5 %
100.0
%
Electron excitation
99.5 %
0.0 %
already in CHIANTISlide12
SummaryStudying level populations along isoelectronic sequences is a good way of assessing atomic data accuracy.The method focuses attention on those atomic data that most influence the populations (and hence emissivities) of the ions.For Z ≥ 20 large atomic models are critical for modeling weak ground-level excitations.Software will be distributed with CHIANTI 8.
Dr Peter Young (GMU)http://pyoung.org/talks/aas201412See online talk for He- to Na-like sequences(http://pyoung.org/talks/aas2014)Slide13
C-like 2p4 levelsMinor elements show large differences (e.g., P X vs. S XI)Dr Peter Young (GMU)http://pyoung.org/talks/aas2014
13Slide14
P X & S XI: 2p4 1S0 levelFactor 10 difference in level population for 2p4 1S0
level – why?Error for P X oscillator strengthDr Peter Young (GMU)http://pyoung.org/talks/aas201414Transition
Zhang & Sampson
(1996)
CHIANTI
2s 2p
3
1
P
1
- 2p
4
1
S
0
0.388
3.49
The Z&S data have been replaced for the abundant elements, but not the minor elementsSlide15
Extra slide: Be-like dataDr Peter Young (GMU)http://pyoung.org/talks/aas2014
15Slide16
Extra slide: B-like dataDr Peter Young (GMU)http://pyoung.org/talks/aas2014
16Slide17
Extra slide: N-like dataDr Peter Young (GMU)http://pyoung.org/talks/aas2014
17Slide18
Extra slide: O-like dataDr Peter Young (GMU)http://pyoung.org/talks/aas2014
18Slide19
Extra slide: F-like dataDr Peter Young (GMU)http://pyoung.org/talks/aas2014
19Slide20
Extra slide: Ne-like dataDr Peter Young (GMU)http://pyoung.org/talks/aas2014
20Slide21
Extra slide: Na-like dataDr Peter Young (GMU)http://pyoung.org/talks/aas2014
21