Exploring NEI based on AtomDB 3.0

Li Ji (纪丽) [email protected]

中国科学院暗物质与空间天文重点实验室!Key Laboratory of Dark Matter and Space Astronomy!

Chinese Academy of Sciences

Exploring Non-equilibrium Ionization with AtomDB

2014.9.7 AtomDB Workshop, Japan

mailto:[email protected]

-Introduce scientists to high energy astrophysics and multi-wavelength opportunities-Develop knowledge of the XMM-Newton, Chandra, and Suzaku data archives-Train scientists in the techniques required to access and analyse data from X-ray archives-Introduce scientists to X-ray instrumentation applicable to operating future X-ray missions-Train scientists in the use of X-ray analysis software-Provide a general overview of possibilities provided by future space-based missions

PhD students, postdocs and young faculty member from developingcountries in the Asian region, who are either in X-ray fields or withemphasis on multi-wavelength opportunities, are welcome to apply for the workshop. The total number of attendants, however, will be limited to 30.

Deadline for registration: June 20th, 2013Deadline for financial support request: June 10th, 2013For more information, please visit: http://cospar2013xos.csp.escience.cn

High-Energy Astrophysics: An Advanced School for Asian Astronomers

- Basic theoretical concepts of High Energy Astrophysics- Statistics: basic concepts and applications- High Energy Satellites: past, present and future- Software for high-energy data analysis:

installation, usage, and basic scripting- High energy data analysis

- Accessing archival databases for spectral, timing & imaging data- Reduction and analysis of spectroscopic, timing & imaging data- Simultaneous analysis of data from multiple detectors & wavebands

- Basics of writing a scientific paper and an observational proposal

Topics of the Workshop

Workshop Objectives

Target Participants

Lectures

SOC Members

LOC Members

Li Ji, PMO, CAS, (Chair)Yizhong Fan, PMO, CASSiming Liu, PMO, CASHaibing Zhao, PMO, CASBing Jiang, Nanjing UniversityJingying Wang, Shanghai Jiaotong Univ.

Shijun Lei, PMO, CASXin Zhou, PMO, CASShuinai Zhang, PMO, CASYueyue Li, PMO, CASHongling Chen, PMO, CASYi Song, PMO, CASBenyun Hong, PMO, CAS

- Keith Arnaud, CRESST, NASA, USA- Tomaso Belloni, INAF-OAB, Italy- Carlos Gabriel, ESA, Spain- Matteo Guainazzi, ESA, Spain- Mariano Mendez, KAI, The Netherlands- Michael Nowak, MIT, USA- Randall Smith, SAO/CfA, USA- Yukikatsu Terada, Saitama Univ., Japan- Peter Willmore, Univ. of Birmingham, UK- Yongquan Xue, USTC, China

A COSPAR Capacity Building Workshop

Sept. 2nd - 13th, 2013, Xuyi, Jiangsu, China

Mariano Mendez, KAI, The Netherlands- COSPAR (Chair)

Carlos Gabriel, ESAC, ESA, Spain- COSPAR (Co-Chair & Coordinator)

Alvaro Gimenez, ESA, SpainKwong Sang Cheng, Hongkong UniversityTiepei Li, Tsinghua UniversityShuangnan Zhang, High Energy Institution, CASWeimin Yuan, NAO, CASQingzhong Liu, PMO, CASQiusheng Gu, Nanjing UniversityXiangdong Li, Nanjing UniversityYang Chen, Nanjing UniversityHaiguang Xu, Shanghai Jiao Tong UniversityJin Chang, PMO, CASDaming, Wei, PMO, CASWeiqun, Gan, PMO, CASIan Corbett, IAU, FranceNeil Gehrels, Goddard Space Flight Center, USAHans Haubold, UN/OOSARobert Missotten, UNESCO, Paris, FranceRandall Smith, SAO/CfATakaya Ohashi Tokyo Metropolitan University, JapanPeter Willmore, University of Birmingham, UK-COSPAR










Workshop Objectives

Target Participants

Lectures

SOC Members

LOC Members


















Workshop Objectives

Target Participants

Lectures

SOC Members

LOC Members


















Workshop Objectives

Target Participants

Lectures

SOC Members

LOC Members


















Workshop Objectives

Target Participants

Lectures

SOC Members

LOC Members


















Workshop Objectives

Target Participants

Lectures

SOC Members

LOC Members









X-ray Spectroscopy Group in PMO

Research faculties: Lei, S.J. ; Zhou, X.; Yan, J.Z.;!Postdocs: Zhang, S.N.; Yan, S.P. Graduate students: Niu, S.; Wang, Q.; Fu, X.D.;Zhang, B. (joint) Undergraduate students: Yu, C.; Ji, Z.; Zhang, Z.; Cheng, H.


Open for visiting Scholars/Postdocs/PhDs

Outlines

- Where and how does NEI happen (Recap)? !- What’s the requirements if considering NEI in your plasmas? !- What are NEI applications in the different plasmas ? - What do we need in the future?


Where and How does NEI happen? Recap


Astrophysical Plasma

General Ionization balance:

Plasma: - Collisional ionization plasma: - Photo-ionization plasma: ! - Mixed

kTe << Ionization energy of plasma ions

kTe ⇠ Ionization energy of plasma ions

≠0 (NEI) - delayed ionisation (recombing plasma) - delayed recombination ( ionizing plasma)

= 0 (Equilibrium)

1ne

dNzdt = Nz�1Sz�1 �Nz(Sz + �z) + Nz+1�z+1


I��

� �

�+��,�� -��

��.�� /��

"��0��1�� +��/

�2��34"��0��

52��34�

%��/��

��

6�/�� 7

%��/+��*

�8��.��/��

.��/ #�� 7��

6��7��

�e Te

��

Credit: H. Böhringer 1997

Basic Structure of Plasma Models

AtomDB: v1.3.1 (2001), v 2.0 (2010 released) v3.0 (2014)


Hot Plasmas

Eta Car

Corcoran et al. 2007

Sgr A*

M104 Sombrero

1-8 keV

HST MICMOS Near-IR

Arches cluster and its vicinity

Figer et al. 2000

D25 XMM-EPIC pn

NGC253

Breitschwerdt 2002

Baganof et al 2003

Li et al. 2009

Astrophysical Plasmas


- Solar flares - X-ray binaries - Sgr A* - Super Stellar Clusters - SNRs - Galaxies - Cluster of galaxies - ICM & IGM

Hot Plasmas


XMM-Newton

Chandra X-ray Observatory

a high resolution mirrortwo imaging detectorstwo sets of transmission gratings

Instruments

Important features: an order of magnitude in spatial resolution 0.5 arcsec good sensitivity from 0.1 – 10keV high spectral resolution

HETGs: High Energy Transmission Grating Spectrometer, 0.4 to 10keV

High Energy ObservatoriesInstrument: a high resolution mirror,two imaging detectors; two sets of transmission gratings Important features: spatial resolution 0.5 arcsec good sensitivity from 0.1-10 keV high spectral resolution !HETGs: high energy transmission grating spectrometers 0.4-10keV


What is the theoretical requirements/predictions for NEI plasmas?


74 S.M. Kahn

of photoionized plasmas, Lyman series lines are formed by radiative cascadesassociated with radiative recombination. The line ratios produced by theseprocesses are somewhat different than those associated with collisional exci-tation in collisional plasmas. This is apparent from Fig. 14, where it can beseen that the Ly β to Ly α ratio for O VIII is ∼0.11 for a coronal plasma,and ∼0.14 for a photoionized plasma. Similar enhancements are found forthe higher series line ratios as well.

6.2 He-like Transitions

He-like K-shell lines are among the most important of all in the soft X-ray band. Since the He-like charge state is a tight “closed shell”, this isthe dominant ion species over a wide range in temperature, particularly incoronal plasmas. In addition, as explained below, these lines exhibit strongsensitivity to electron density, temperature, and ionization conditions in theemitting plasma.

The most important K-shell He-like transitions are as follows:

W : 1s2 1S0 – 1s2p 1P1

X: 1s2 1S0 – 1s2p 3P2

Y : 1s2 1S0 – 1s2p 3P1

Z: 1s2 1S0 – 1s2p 3S1

W is an electric dipole transition, also called the resonance transition, andis sometimes designated with the symbol r. X and Y are the so-called in-tercombination lines. These are usually blended (especially for the lower-Zelements), and are collectively designated with the symbol i. Z is the forbid-den line, often designated by the symbol f . It is a relativistic magnetic dipoletransition, with a very low radiative decay rate.

The temperature sensitivity of these lines arises as follows [14–16]: SinceW is an electric dipole transition, the collision strength for collisional excita-tion of this line includes important contributions from higher order terms inthe partial wave expansion, and thus continues to increase with energy abovethreshold. By contrast, X and Z are electric dipole forbidden. The dominantterm in the excitation collision strength for these transitions involves elec-tron exchange. Therefore, their excitation collision strengths drop off stronglywith energy above threshold, whereas Y remains relatively constant. As a re-sult, the line ratio: G = (X + Y + Z)/W is a decreasing function of electrontemperature.

The density sensitivity comes from the fact that the 3S1 level can becollisionally excited to the 3P levels. At high electron density, that processsuccessfully competes with radiative decay of the forbidden line. Therefore,the ratio R = Z/(X + Y ) drops off above a critical density, nc. The criticaldensity depends strongly on Z. For C V, nc ∼ 109 cm−3, while for Si XIII,nc ∼ 1013 cm−3.

He-like Transitions- the most important K-shell He-like transitions

- resonance line (r)

- intercombination lines (i)

- forbidden lines (f)

- temperature sensitive: G ratio

74 S.M. Kahn





W : 1s2 1S0 – 1s2p 1P1

X: 1s2 1S0 – 1s2p 3P2

Y : 1s2 1S0 – 1s2p 3P1

Z: 1s2 1S0 – 1s2p 3S1




- density sensitive: R ratio

74 S.M. Kahn





W : 1s2 1S0 – 1s2p 1P1

X: 1s2 1S0 – 1s2p 3P2

Y : 1s2 1S0 – 1s2p 3P1

Z: 1s2 1S0 – 1s2p 3S1




The Astrophysical Journal, 756:128 (11pp), 2012 September 10 Foster et al.

T = 3.56x10 Kn = 5.8x10 cm

e

e-3

6

0.0

0.2

0.4

0.6

0.8

9.10 9.15 9.20 9.25 9.30 9.35 9.40 9.45 9.50

0.0

0.4

0.8

Fe X

VII

Ne

X, n

=51

Na

X, 3

p1s

Ne

X, n

=61

Ne

X, n

=71

Mg

XI,

DR

Ne

X, n

=81

Ne

X, n

=91

Ne

X, n

=10

1M

g X

I, D

R

Mg XIw, x+y, z

Wavelength [A]

Nor

m. I

nten

sity

Nor

m. C

ount

s

Figure 6. Top: summed Chandra HETG spectrum of TW Hydrae (Brickhouseet al. 2010) in the Mg xi bandpass. Bottom: a simulated AtomDB spectrum withGaussian broadening applied to each line. Contributions from Ne x are shadedin red.(A color version of this figure is available in the online journal.)

10 6 10 7 10 8

Te [K ]

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

G r

atio

CONeSiArFe

v2.0.2v1.3.1

Figure 7. Temperature-sensitive G-ratio for selected He-like ions, for both theold (v1.3.1, dashed) and new (v2.0.2, solid) AtomDB.(A color version of this figure is available in the online journal.)

These values are taken from Model A in Brickhouse et al.(2010), as are the plasma conditions of Te = 3.58 × 106 K,ne = 5.8 × 1012 cm−3. The inclusion of the n > 5 levels leadsto several extra lines being apparent in the spectrum, potentiallyaffecting line ratio estimates depending on the resolution of theinstrument in use.

4.4. Diagnostic Line Ratios

4.4.1. He-like Ratios

The effect on the G-ratios (Section 3.4) of the new data usedin AtomDB v2.0.2 can be seen in Figure 7. For most ions theG-ratio substantially increases, in some cases leading to a dou-bling of the implied temperature. At the peak temperatures foreach ion, this effect is attributed largely to an increase in thecollisional excitation rate to the forbidden and intercombina-tion transition upper levels: this has led to, for example, a 30%increase in both the forbidden and intercombination line emis-sion for Si xiii. At slightly higher temperatures, above which the

106 107

Te[Lyα /R]

106

107

Te [G

]

Magnesium

AtomDB 1.3.1

106 107

Te[Lyα /R]

106

107

Te [G

]

Magnesium

AtomDB 2.0.2

106 107

Te[Lyα /R]

106

107

Te [G

]

Oxygen

AtomDB 1.3.1

106 107

Te[Lyα /R]

106

107

Te [G

]

Oxygen

AtomDB 2.0.2

Figure 8. Comparison of the temperatures implied by the measured G-ratiosand Lyα/resonance lines for Chandra MEG observation data from Testa et al.(2004), using AtomDB v1.3.1 data (as in the original paper) and the newAtomDB v2.0.2 data. Error bars are based on the error in the line fluxes fromTesta et al. (2004); X-axis errors are very small due to the very strong temperaturedependence of the Lyα/w ratio.(A color version of this figure is available in the online journal.)

n = 2 → 1 complex emission has peaked, the recombinationand cascade processes have the most significant effect.

Comparisons of the temperature implied by the G-ratiowith that implied by the ratio of the Lyα/w, where w is theHe-like resonance line, are sometimes used to investigate plasmaionization states. The two line ratios measure slightly differenttemperatures. The G-ratio measures directly the electron tem-perature of the plasma, due to the fast electron collision rates.The Lyα/w ratio, however, primarily depends on the ionizationbalance, and therefore can be used to indicate the ionizationtemperature of the plasma. If these two diagnostics disagreeabout the electron temperature, then the plasma is either ove-rionized or underionized; this is possible since ionization andrecombination occur on much slower timescales than collisionalexcitation. For a single temperature plasma in ionization equi-librium they will be equal; however, in an object with a broadthermal distribution, the Lyα/w ratio can be elevated due to Lyαemission from warmer plasma.

Ness et al. (2003) encountered this problem when comparingthe G-ratio for Ne ix with the emission measure obtained fromiron L-shell ions in Capella: the emission measure peaks ataround 6 MK, but the G-ratio implied a Te of around 2 MK. Testaet al. (2004) noted that in Chandra observations of a variety ofstars, the temperature was consistently underestimated by theG-ratios, sometimes by up to a factor of four. In Figure 8, werepeat this comparison, using the data from AtomDB v2.0.2,and the line fluxes from Tables 6 and 7 of their paper. Usingthe AtomDB v1.3.1 data, we have been unable to reproducethe exact temperatures from their paper, but the same trend isclearly evident.

Testa et al. (2004) and Ness et al. (2003) discuss the causes ofsome of these discrepancies. The differential emission measure

9

The Astrophysical Journal, 756:128 (11pp), 2012 September 10 Foster et al.

AtomDB v2.0.2AtomDB v1.3.1

Smith+2009 (full)Smith+2009 (No Recomb)

Ne IX

n = 1cm e−3

106

107

Temperature (K)

2

3

4

5

6

R R

atio

Figure 9. Density diagnostic R-ratio from AtomDB v1.3.1 (dash dotted), theAtomDB v2.0.2 data (solid), and the data from Smith et al. (2009) both with(dotted) and without (dashed) recombination.(A color version of this figure is available in the online journal.)

(DEM) of the stars in the survey is generally peaked around107 K (see references in Testa et al. 2004). The temperaturesof peak emissivity of O vii and Mg xii are ≈2 × 106 K and≈6.4×106 K, respectively. By 107 K, the emissivity of the O viihas dropped two orders of magnitude from its peak. Thereforewe would expect the G-ratio from oxygen to be slightly higherthan the peak emissivity at 2 × 106 K, but not as high as 107 Kdue to the lack of O vii emission at this temperature. The Mg xiG-ratio should give a temperature closer to the peak of theDEM at 107 K as the emissivity of this ion is still high. Instead,they found that the G-ratio in both cases gave temperatureswhich were below the temperatures of peak emissivity, whichsuggested a problem in the atomic data.

With the AtomDB version 2.0 data, we find that the G-rationow indicates temperatures which lie on average at or above thepeak emissivity temperature for each ion (some spread remainsdue to the different DEMs of different objects). In addition,the discrepancy between the two temperature measurementmethods has been reduced from 60% to around 35%.

Smith et al. (2009) also investigated the G- and R-ratiosfor the case of Ne ix. They used AtomDB v1.3.1 data, how-ever they updated it to include the more recent collisionaldata of Chen et al. (2006) (discussed in Section 3.4) andthe recombination data from the Badnell (2006a) collection—the same data included in the release of AtomDB v2.0.2. Theauthors found a significant variation with temperature in theR-ratio, which is the ratio of the forbidden to the intercom-bination lines of He-like ions. This temperature dependence isunusual: the G-ratio is typically a widely used density diagnosticwith little temperature variation. We do not find any temperaturedependence here, despite using almost identical data (the smalldifferences in the collisional excitation calculations have almostno effect). This was traced to an error in the Smith et al. (2009)paper when connecting the energy levels in the recombinationdata to the excitation data. Figure 9 shows the corrected R-ratioas a function of temperature.

5. SUMMARY AND FUTURE PLANS

We have presented the latest version of AtomDB, the firstmajor update since 2001. Nearly every single piece of data in

the database has been updated, with many new ions added.This is the result of a comprehensive evaluation of the previousdata, and assessment of its replacement, addressing many of theknown issues in the previous version. New recombination datasignificantly alter the ionization balance, and therefore the emis-sivities, of many lines; new collisional data for H- and He-likeions make significant improvements to common temperaturediagnostics.

The new data are now available online at www.atomdb.org,and are also available through spectral fitting packages such asXSPEC (Arnaud 1996), ISIS (Houck & Denicola 2000), andSherpa (Freeman 2001).3

There are several improvements already planned for the nextrelease of AtomDB. We will include final state resolved DRrates for the remaining ions in the database for which the dataexist. We will largely target non-equilibrium ionization plasmas,with new inner-shell excitation data, as well as fluorescence linedata. Work is also ongoing to document and release the APECcollisional ionization code, which will allow users to generatethe non-equilibrium higher density plasma models.

We thank Paola Testa for discussion involving the linediagnostic issues. The authors gratefully acknowledge fundingfrom NASA ADP grant NNX09AC71G.

REFERENCES

Ait-Tahar, S., Grant, I., & Norrington, P. 1996, Phys. Rev. A, 54, 3984Altun, Z., Yumak, A., Badnell, N. R., Colgan, J., & Pindzola, M. S. 2004, A&A,

420, 775Altun, Z., Yumak, A., Badnell, N. R., Loch, S. D., & Pindzola, M. S. 2006, A&A,

447, 1165Altun, Z., Yumak, A., Yavuz, I., et al. 2007, A&A, 474, 1051Anders, E., & Grevesse, N. 1989, Geochim. Cosmochim. Acta, 53, 197Anderson, H., Ballance, C. P., Badnell, N. R., & Summers, H. P. 2002, J. Phys.

B: At. Mol. Opt. Phys., 35, 1613Arnaud, K. A. 1996, in ASP Conf. Ser. 101, Astronomical Data Analysis

Software and Systems V, ed. G. H. Jacoby & J. Barnes (San Francisco,CA: ASP), 17

Arnaud, M., & Raymond, J. 1992, ApJ, 398, 394Badnell, N. R. 1986, J. Phys. B: At. Mol. Phys., 19, 3827Badnell, N. R. 2006a,Dept. Phys. Univ. Strathclyde, http://amdpp.phys.strath.ac.

uk/tamoc/DATA/Badnell, N. R. 2006b, A&A, 447, 389Badnell, N. R., & Griffin, D. C. 2001, J. Phys. B: At. Mol. Opt. Phys., 34, 681Badnell, N. R., Griffin, D. C., & Mitnik, D. M. 2001, J. Phys. B: At. Mol. Opt.

Phys., 34, 5071Ballance, C. P., Badnell, N. R., & Smyth, E. S. 2003, J. Phys. B: At. Mol. Opt.

Phys., 36, 3707Bar-Shalom, A., Klapisch, M., & Oreg, J. 2001, J. Quant. Spectrosc. Radiat.

Transfer, 71, 169Bartschat, K., Hudson, E., Scott, M., Burke, P., & Burke, V. 1996, Phys. Rev.

A, 54, R998Bautista, M. A., & Badnell, N. R. 2007, A&A, 466, 755Beiersdorfer, P., Behar, E., Boyce, K. R., et al. 2002, ApJ, 576, L169Boyle, J. J., & Pindzola, M. S. 2005, Many-body Atomic Physics (Cambridge:

Cambridge Univ. Press), 428Brickhouse, N. S., Cranmer, S. R., Dupree, A. K., Luna, G. J. M., & Wolk, S.

2010, ApJ, 710, 1835Brickhouse, N. S., Raymond, J. C., & Smith, B. W. 1995, ApJS, 97, 551Brickhouse, N. S., & Schmelz, J. T. 2006, ApJ, 636, L53Brown, G. V., Beiersdorfer, P., Chen, H., Chen, M. H., & Reed, K. J. 2001, ApJ,

557, L75Brown, G. V., Beiersdorfer, P., Liedahl, D. A., Widmann, K., & Kahn, S. M.

1998, ApJ, 502, 1015Bryans, P., Badnell, N. R., Gorczyca, T. W., et al. 2006, ApJS, 167, 343

3 After the release of v2.0.0, errors were discovered affecting the radiativerecombination continuum and the autoionization rates. This led to the releaseof v2.0.1 and v2.0.2 of AtomDB. All data in this paper refer to the correctedv2.0.2 data.

10

Foster et al. 2012 ApJ2014.9.7 AtomDB Workshop, Japan

��

DR & RR cascades

Credit: A.Foster

Dielectronic /Radiative Recombination Cascades

Important in some Non-equilibrium Ionization scenarios


CVCVI OVII

OVIII

Dielectronic /Radiative Recombination Cascades An Illustration

r=2pc

An adiabatic expanding star cluster wind

Ji et al. 2006 MNRAS

Continuum (black) Lines: - with (red) - without (blue) cascades

T=4e5 K

NEI code and its application 501

Figure 4. CIE ion fractions of our model (red solid line) and Mozz98’s model (black dotted line) for elements C, N, O and Fe as marked in individual panels.

3.4 Non-equilibrium ionization calculations

In general, the dynamics and ionization of a plasma are entangled.The ionization structure, through its heating/cooling effects, can af-fect the temperature evolution. However, in a cluster wind, radiativecooling can typically be neglected and the dynamics and ioniza-tion of the wind can be separated. With the mass input having acertain spatial distribution in our model, and assuming the newlyinjected gas to be neutral, the cluster wind number density of theneutral ion NZ ,0 for a given element Z evolves in a cluster windas

1r 2

dNZ ,0ur 2

dr=

Zm(r )m Z

− ne NZ ,0 SZ ,0 + ne NZ ,1αZ ,1, (13)

with Z and mZ being the abundance and mass of element Z.The number density of other ions (NZ ,z>0) of the same element

follows from1ne

dNZ ,z>0

dt= NZ ,z−1 SZ ,z−1 − NZ ,z(SZ ,z + αZ ,z) + NZ ,z+1αZ ,z+1,

(14)

where SZ ,z and αZ ,z are the total ionization (z → z + 1) and recom-bination (z → z − 1) rate coefficients (in cm3 s−1) of ion Z+z . Ourinterest here is in highly ionized ions. Therefore, the exact ioniza-tion state of the injected gas is not important as long as it is cold orwarm (i.e. T ! 104 K).

For each element, we simultaneously solve these differentialequations using IDL code LSODE which adaptively solves stiff andnon-stiff systems of ordinary differential equations.

In an evolving plasma, a characteristic time-scale to approach aCIE state is τCIE ≈ [ne(SZ ,z + αZ ,z)]−1 (Mewe 1997). The CIE ap-proximation requires the cooling/heating time be much longer thanτCIE. As mentioned earlier, CIE is only a special case. If experienc-ing a rapid heating or cooling process such as thermal instability,shock, or rapid expansion, a plasma may be underionized or overion-ized, compared to the CIE result (Dopita, Sutherland & Schweiger2002).

As an illustration, we consider a toy model for an adiabati-cally expanding stellar cluster wind with a constant mass input rateM ∼ 3 × 10−5 M⊙ yr−1 and a constant velocity v = 1000 km s−1.We assume that the wind is injected at an initial radius of r0 =0.3 pc and is heated up to a CIE state with an equilibrium tem-perature T0 = 5 × 106 K. As the wind expands adiabatically,its temperature drops as T = T0(r/r0)−4/3. Fig. 5 compares theCIE and non-CIE spectra at r = 1 pc, where T ∼1 × 106 K (left-hand panel) and at r = 2 pc, where T ∼ 4 × 105 K (right-handpanel). The difference is significant: in the lower energy part, theline intensities in the non-CIE spectrum are weaker than those inthe CIE case, while in the upper energy part sawtooth structuresare present in the non-CIE spectra; all these are due to delayedrecombinations.

C⃝ 2006 The Authors. Journal compilation C⃝ 2006 RAS, MNRAS 372, 497–509





1r 2

dNZ ,0ur 2

dr=

Zm(r )m Z

− ne NZ ,0 SZ ,0 + ne NZ ,1αZ ,1, (13)


follows from1ne

dNZ ,z>0


(14)










1r 2

dNZ ,0ur 2

dr=

Zm(r )m Z

− ne NZ ,0 SZ ,0 + ne NZ ,1αZ ,1, (13)


follows from1ne

dNZ ,z>0


(14)










1r 2

dNZ ,0ur 2

dr=

Zm(r )m Z

− ne NZ ,0 SZ ,0 + ne NZ ,1αZ ,1, (13)


follows from1ne

dNZ ,z>0


(14)










1r 2

dNZ ,0ur 2

dr=

Zm(r )m Z

− ne NZ ,0 SZ ,0 + ne NZ ,1αZ ,1, (13)


follows from1ne

dNZ ,z>0


(14)










1r 2

dNZ ,0ur 2

dr=

Zm(r )m Z

− ne NZ ,0 SZ ,0 + ne NZ ,1αZ ,1, (13)


follows from1ne

dNZ ,z>0


(14)







Inner-shell Processes

e.g. Li-like ions make contribution to the wavelength region of He-like

– 4 –

n t(cm s )−3 −1e n t(cm s )−3 −1

e e−3 −1n t(cm s ) en t(cm s )−3 −1

e−3 −1n t(cm s ) e

−3 −1n t(cm s )

OVII K / OVIII Lyαα

81010

41020

10−2

1012

1010

106

10

T (

keV

)

T (

keV

)

T (

keV

)

a b c

T (

keV

)

T (

keV

)

T (

keV

)

ba c

SiXIII K / SiXIV Lyαα

Fig. 3.— Line diagnostics based on vpshock in XSPEC, assuming [net]low = 0 based ondifferent versions of AtomDB with (a) and without (b,c) innershell processes: a— upcoming

version; b — v2.0; c — v1.3

parallel shock plasma model (vpshock in XSPEC). Note Li-like ions make contributions tothe wavelength region of He-like Kα due to inner-shell processes. As expected, such line ra-tios are enhanced when the temperature is lower and ionization time is shorter. This Li-like

data will be incorporated in a future release of AtomDB.

We gratefully acknowledge support for this work form NASA ADP grant NNX09AC71Gand from Chandra HETG group for database realization in ISIS.

REFERENCES

Badnell, N.R., 1997, JPhB, 30, 1

Badnell, N.R., O’Mullane, M.G., Summers, H.P., et al., 2003, A&A, 406, 1151

Foster, A. R., Smith, R.K., Ji, L. & Brickhouse, N.S., 2011, this proceedings

Griffen J., 1998, Phys. B: At Mol. Opt. Phys. 31, 3713

Ji, L., Wang, Q.D. & Kwan, J., 2006, MNRAS, 372, 497

Reeves, K.K, Warren, H.P. & Forbes, T.G., 2007, ApJ, 668, 1210

This preprint was prepared with the AAS LATEX macros v5.2.

�, e� + I� > I⇤ + 2e�� > I+ + e�, �

Inner-shell Ionization:

Inner-shell Excitation

Electrons from an “inner” shell may be collisionally excitedas well. In some cases this is unremarkable; in Fe XIX,

2s2 2p4 3P2 # 2s 2p5 3P0

could be called a innershell excitation. It decays with a106.3Å photon but it isn’t incredibly significant. However,some transitions are more important:

1s2 2s1 2S1/2 # 1s 2s2 2S1/2

This excitation in Li-like ions creates a photon called the qline, which appears between the resonance and forbiddenlines in the same ion’s He-like system. Seeing this lineshows that there is a noticeable amount of Li-like ion in theplasma.

Inner-shell Excitation:

For example: FeI - FeXVI : emit ~6.4 KeV photons


plane-parallel shocked plasma (vpshock)

Line Diagnostics affected by Inner-shell Process

a: AtomDB 2.1 b: AtomDB 2.0 c: AtomDB 1.3.1

with (a) & without (b,c) Inner-shell process

!

"##$#%&'(

"##$#%&'(

�!"#

#"#

!"#

$"#

%"#

&"#

"#"#

"!"#

)

�*'++#,###-#*'+++#./�

&0##1$23###4###(�5# �6&0##1$23###4###(�5 �6

– 4 –

n t(cm s )−3 −1e n t(cm s )−3 −1

e e−3 −1n t(cm s ) en t(cm s )−3 −1

e−3 −1n t(cm s ) e

−3 −1n t(cm s )

OVII K / OVIII Lyαα

81010

41020

10−2

1012

1010

106

10

T (

keV

)

T (

keV

)

T (

keV

)

a b c

T (

keV

)

T (

keV

)

T (

keV

)

ba c

SiXIII K / SiXIV Lyαα

Fig. 3.— Line diagnostics based on vpshock in XSPEC, assuming [net]low = 0 based ondifferent versions of AtomDB with (a) and without (b,c) innershell processes: a— upcoming

version; b — v2.0; c — v1.3

parallel shock plasma model (vpshock in XSPEC). Note Li-like ions make contributions tothe wavelength region of He-like Kα due to inner-shell processes. As expected, such line ra-tios are enhanced when the temperature is lower and ionization time is shorter. This Li-like

data will be incorporated in a future release of AtomDB.

We gratefully acknowledge support for this work form NASA ADP grant NNX09AC71Gand from Chandra HETG group for database realization in ISIS.

REFERENCES

Badnell, N.R., 1997, JPhB, 30, 1

Badnell, N.R., O’Mullane, M.G., Summers, H.P., et al., 2003, A&A, 406, 1151

Foster, A. R., Smith, R.K., Ji, L. & Brickhouse, N.S., 2011, this proceedings

Griffen J., 1998, Phys. B: At Mol. Opt. Phys. 31, 3713

Ji, L., Wang, Q.D. & Kwan, J., 2006, MNRAS, 372, 497

Reeves, K.K, Warren, H.P. & Forbes, T.G., 2007, ApJ, 668, 1210

This preprint was prepared with the AAS LATEX macros v5.2.

Ji et al. 2010 NASA-Law Workshop



Radiative Cooling in Hot PlasmasBased on AtomDB

Credit: Zhou, X. et al. 2014 to be submitted

- CIE: !!- Ionizing & recombing plasmas: Te = 104 � 108 Knet = 1� 1013 cm�3s

ne = 0.1� 1013 cm�3

Te = 104 � 109 K

CIE Cooling Curve



Radiative Cooling of Non-equilibrium Ionization Plasmas Based on AtomDBv2.xXin Zhou1, Li Ji1, Adam R. Foster2, Randall K. Smith2, Nancy S. Brickhouse2, Hiroya Yamaguchi2, Ji Yang1

1Purple Mountain Observatory, CAS2Harvard-Smithsonian Astrophysical Observatory, USA

The detail ionization state is difficult to know from observations, collisional ionization equillibrium (CIE) or simple parameterized ionizing assumption are usually applied in modeling. But such assumption doesn’t work in some respects, e.g. for studying the overionized plasma which is recently confirmed in supernova remnants. We performed this work to study the different radiative cooling rates in CIE and non-equillibrium ionization (NEI) cases in detail, which is crucial in modeling the evolution of plasma in realistic cases. This work is based on the most developed atomic database, AtomDB v2.0.2 (see Foster et al. 2012, ApJ, 756, 128), and applied the companied APEC program for calculations. We focus on the hot optically thin gas, such as the diffuse interstellar medium. We present our primary results of CIE plasma in a density range of 1 cm-3 to 1014 cm-3 and a temperature range of 104 K to 109 K, and for NEI plasma, both ionizing and recombining cases with the initial ionization temperatures of 104 K and 108 K, respectively and with the ionization timescales from 1 cm-3 s to 1013 cm-3 s. There are 30 elements included in the calculation, from hydrogen to zinc. Detailed comparison between CIE and NEI plasma are shown. Further implication from such comparison are discussed.

CIE ionizing recombining

Acknowledgement: We gratefully acknowledge support for this work by NSFC grant, China Postdoctoral Science Foundation grant, and 100 Talents Program of CAS

abstract

Fe ionization balance

-ionizing: the ions are ion ized gradual ly and approaching CIE at net~1012 cm-3 s-recombining: the ionization levels are mixed together at net~1011 cm-3 s

Cooling efficiency

-ionizing: the radiative cooling rate is higher than CIE plasma and gradually approach the CIE case -recombining: the radiative cooling rate is lower than CIE plasma and doesn’t change much for different temperatures at beginning and quickly approaches the CIE case at the late stage

cooling function

-line emission dominates in most of the cases- i on i z ing : two-pho ton emission dominates the continuum for net<3*1010 cm-3 s and bremsstrahlung dominates for net above- r e c o m b i n i n g : R R C emission is comparable to line emission for net<1011 cm-3 s

spectrum

-ionizing: emission lines become weaker and weaker in the ionizing process-recombining: the RRC features of Fe is prominent at around 9 keV and doesn’t change much for net<1011 cm-3 s

Friday, April 5, 13



Ionizing PlasmaTi = 104 KTe = 107 Kne = 0.1cm�3








abstract



Cooling efficiency


cooling function


spectrum


Friday, April 5, 13






abstract



Cooling efficiency


cooling function


spectrum


Friday, April 5, 13




Recombing PlasmaT0 = 108KTe = 106Kne = 0.1cm�3


What are NEI applications?


Super Stellar Clusters

Figer et al. 2000, Wang et al. 2006

Arches cluster and its vicinity

1-8 keV

HST MICMOS Near-IR

mass and energy loss from massive stars energize ISM by their winds (stellar cluster wind)

estimation of mass and energy injections suffers significant uncertainties

lie in extra-galactic star-forming galaxies (e.g. M82) or in local regions (e.g. NGC3603, R136 in 30 Doradus, Arches cluster)

super stellar clusters (SSCs): super stellar clusters (SSCs):

based on collisional ionization equilibrium(CIE) assumption


Observation: Stellar Cluster WindNGC3603

1 arcmin --- 2.036pc

Chandra ACIS-I 0.5-8.0keV

Db

Da

Ji et al. 2006

Application I


2012 Workshop on Ecosystems of Galaxies

Stellar Cluster WindApplications I

Table 4.3. Joint spectral fit resultsa

Parameter Model

NH (1022 cm−2) 0.81+0.06−0.05

Cluster wind componentrsc (pc) 0.68 (fixed)

Abundance (Z⊙) 0.21+0.12−0.11

LogM0 (M⊙ yr) −3.37+0.42−0.24

V∞ (1 × 103 km s−1) 1.53 (> 0.50)Point-like source component

kT (keV) 3.79+0.41−0.33

K1b (10−4 photons cm−2) 3.63+0.77

−1.56

K2c (10−4 photons cm−2) 7.76+1.38

−1.73

K3d (10−3 photons cm−2) 3.26+0.22

−0.19

χ2/d.o.f. 301.3/457

aNote: all error bars are at the 90% confidencebK1 is the MEKAL normalization of the unde-

tected point sources in Region Da.cK2 is same as K1, but in Region Db.

dK3 refers to the MEKAL normalization of the de-tected point sources.

68


Da

Db

Figure 16. Chandra ACIS-I observation of NGC 3603. Left-hand panel: the 0.5–8.0 keV intensity contours (at 52, 58, 70, 94, 130, 178, 238, 310, 394 and 490 ×10−4 counts s−1 arcmin−2) overlaid on an optical image. Right-hand panel: a close-up of the central cluster. The image shows the ACIS-I event distribution.Point-like X-ray sources are marked individually with small circles of 2× the PSF 90 per cent EER. The two annulii, Da ([19, 30] arcsec) and Db ([30, 83] arcsec),are used to extract the spectra of the diffuse emission.

primarily discrete X-ray sources, although they also character-ized the remaining ‘diffuse’ emission with a one-temperatureCIE thermal plasma model. Here, we concentrate on the diffuseemission.

We reprocess the ACIS-I event data, using the Chandra Inter-active Analysis of Observations software package (CIAO, version3.2). This reprocess, including the exclusion of time intervals withbackground flares, results in a net 47.4-ks exposure (live time) forsubsequent analysis.

The superb spatial resolution of the data allows for the detec-tion and removal of much of the X-ray contribution from individualstars. We search for X-ray sources in the 0.5–2, 2–8 and 0.5–8 keVbands by using a combination of three source detection and analysisalgorithms: wavelet, sliding-box and maximum likelihood centroidfitting (Wang 2004). Source count rates (CRs) are estimated withthe 90 per cent energy-encircled radius (EER) of the point spreadfunction (PSF). The accepted sources all have the local false detec-tion probability P ! 10−6. The right-hand panel of Fig. 16 shows thedetected point sources within the cluster field. The source detectionin the inner ∼20 arcsec radius of the cluster is highly incompletedue to severe source confusion.

To analyse the remaining ‘diffuse’ emission from the cluster,we remove the detected source regions from the data. For eachfaint source with a CR ! 0.01 counts s−1, we exclude a region oftwice the 90 per cent EER, which increases with off-axis angle. Foreach source with CR > 0.01 counts s−1, an additional factor of 1 +log (CR/0.01) is multiplied to the source-removal radius to furtherminimize the residual contribution from the PSF wings of relativelybright sources.

We extract the spectra of the diffuse X-ray emission separatelyfrom the two annular regions, Da and Db (divided to contain roughlythe same number of counts), as well as an accumulated point sourcespectrum from the combined region (Da+Db). An additional spec-trum extracted from the outer annulus between 4.0 and 5.0 arcmin

radii is used as the local background of the diffuse X-ray emission,which in turn are normalized for the background subtraction of thesources. Fig. 17 shows a joint fit of the spectra with our clusterwind model plus a thermal model, which is used to characterize thepoint-like source spectrum. Here, we assume that the detected andundetected point sources have the same spectral shape. We jointly fitthe diffuse spectra and the accumulative spectrum of point sourcestogether. We find that the spectral fit results are insensitive to theassumed stellar distributions of our model and to the exact scale

Figure 17. Joint fit to the spectra of the point-like sources (green) and thetwo diffuse X-ray spectra (red and blue) of NGC 3603. All these spectra aregrouped to contain at least 25 net counts per bin. The point-source spectrumis modelled with an optically thin thermal plasma, while the diffuse X-rayspectral model also contains a joint-fitted residual point-like source contri-bution (dotted histograms) as well as the cluster wind component (dashedline).


- First physical modeling!- “Wind” < 50% (Ji et al. 2006)!- Qian et al. 2014 in-preparation!


dash: CIEsolid: non-CIE

Ji et al. 2006


Solar Flares


020

4060

00.511.5

100 200 300 4000

510

wavelength(A)

05

1015

01

23

100 200 300 400

05101520

wavelength(A)

9 �3Te, ne = 0.97 MK, 0.52 x 10 cm

�3911 MK, 9.5 x 10 cm

�394.5 MK, 8.0 x 10 cm 9 �33.1 MK, 7.5 x 10 cm

�397.4 MK, 6.4 x 10 cm

9 �316 MK, 0.46 x 10 cm

�2

Flux

( 1

0 p

hoto

ns c

m s

)�1�1

4

�2

Flux

( 1

0 p

hoto

ns c

m s

)�1�1

4

20 seconds

50 seconds 100 seconds

250 seconds 500 seconds

Initial Conditions

CIE

NEI NEICIE

!"##$%$$$&$!"###$'(��

)$$*+,-$$$.$$$/�0$ �12 )$$*+,-$$$.$$$/�0$ �1

2 2�0 �1)$$*+,-$$$.$$$/ 2)$$*+,-$$$.$$$/�0$ �1

2�0 �1)$$*+,-$$$.$$$/ 2

�0 �1)$$*+,-$$$.$$$/

�345####$%$$$&$345#"$'( �

!"#"#

$"#%#

"#�%

"#"%

"#"#

"#&

"#

6$$+$72"/

6$$+$72"/

6$$+$72"/

8 9 ,

6$$+$72"/

6$$+$72"/

6$$+$72"/

98 ,

!""#"$%&"""'""(�)""�*+

,

-"$.+/(

01)2

)12+3

41)+3

5

6

,/77"8""9",/777":;��

!""#"$%&"""'""(�)""�*+

)12+3

210+3

01**10

-"$.+/(

01*

,/77""<"9"=

!""#"$%&"""'""(�)""�*+

!""#"$%&"""'""(�)""�*+

� �,-.//"0"""1",/./2"345

6

78*

78)9

:"$;+2(

<8)+=

)89+=

,-.///"">"1"?

,-

78)9*87

)89+=

:"$;+2(

987+=

Upper:� All based on vpshock in XSPEC, assuming [n

e

t]low

= 0. vpshock: Constanttemperature, plane-parallel shock plasma model.

� With(a)/without(b,c) including inner shell ionization (private contact byBorkowski,K.)

� b: Atomdb 2.0.0; c: Atomdb 1.3.1Lower:

Line diagnostics for vpshock with innershell ionization process. Left: O ions;Right: Si ions.

Line Diagnostics Based On Atomic Database With/Without Inner-Shell Ionization

Acknowledgement: We gratefully acknowledge support for this work from NASA ADP grant #NNX09AC71G & from Chandra HETG group for database realization in ISIS

Non-equilibrium Ionization Diagnostics based on AtomDB L. Ji, A.R. Foster, R.K. Smith, N.S. Brickhouse, J. Houck*

Harvard-Smithsonian Center for Astrophysics? MIT Kavli Institute for Astrophysics & Space Research

Energy Input

Mass reservoir at 104 K

See Mariska et al. 1987

0 500 1000

106

107

2!10

65!

106

2!10

7

time (s)

T (K

)

Temperature

0 500 1000

109

1010

time (s)

T (K

)

Density

! !

!"#$!%&'(%)&(!*$%+,-.!"#$%#&'(#)**+#",#-.$-&.&%"/,

6.48!0.02"0.01keV

#!/! .!" #!1.7" .$ % eV

012,3!%&!4,&(!5!6)2!-%7718%$1.%$92!:%3,7!;<#!,.!17=!>??@A!

!""#$%&'$()*++

! !

!"#$!%&'(%)&(!*$%+,-.!"#$%#&'(#)**+#",#-.$-&.&%"/,

6.48!0.02"0.01keV

#!/! .!" #!1.7" .$ % eV

012,3!%&!4,&(!5!6)2!-%7718%$1.%$92!:%3,7!;<#!,.!17=!>??@A!

!""#$%&'$()*++

Database With Fluorescence Lines

Time-dependent Non-equilibrium Ionization

Radiative/Dielectronic Recombination Cascades

! "! #!! #"! $!!

#!�"

#!�%

#!�&

!'!#

!'#

()*+,-./

012)3,4563()12

4+78000

##!! #$!! #&!! #%!!

#!�"

#!�%

#!�&

!'!#

!'#

()*+,-./

012)3,4563()12

4+7000

! $!! %!! 9!! :!! #!!!

#!�"

#!�%

#!�&

!'!#

!'#

()*+,-./

012)3,4563()12

4+77

! $!! %!! 9!!

#!�"

#!�%

#!�&

!'!#

!'#

()*+,-./

012)3,4563()12

4+77000

;!! #!!! ##!! #$!! #&!! #%!!

#!�"

#!�%

#!�&

!'!#

!'#

()*+,-./

012)3,4563()12

4+07

! "!! #!!!

#!�"

#!�%

#!�&

!'!#

!'#

()*+,-./

012)3,4563()12

4+700

! #!! $!! &!! %!! "!!

#!�"

#!�%

#!�&

!'!#

!'#

()*+,-./

012)3,4563()12

4+7708

! "!! #!!!

#!�"

#!�%

#!�&

!'!#

!'#

()*+,-./

012)3,4563()12

4+708

! "! #!! #"! $!!

#!�"

#!�%

#!�&

!'!#

!'#

()*+,-./

012)3,4563()12

4+780

V1.3.1

CHIANTI 6.0V2.0.0

NEI

V1.3.1

CHIANTI 6.0V2.0.0

NEI

V1.3.1

CHIANTI 6.0V2.0.0

NEI

Transition regions surrounding SgrA*

? Fluorescence lines are often seen

? V1.3.1: no fluorescence lines

? V2.0.0: include fluorescence database

— fluorescence yields from

Kasstra & Mewe (1993)

— Part of yields are Modified

by Gorczyca et al. (2003)

? Right: 1Ms Chandra observations

theoretical calculation has included

fluorescence lines

Solar Flare Loop Models

CVI

OVIIOVIII

CVCVI OVII

OVIII

r=1pc

T=1e6 K

r=2pc

T=4e5 K

M ⇠ 3⇥ 10�5 M�yr�1

v = 1000 kms�1

T = T0(r/r0)�4/3

T0 = 5⇥ 106 Kr0 = 0.3 pc

An Adiabatic Expanding Wind

� Fully NEI calculations

� Left: Spectral Comparisons:

— with (red) cascades

— without (blue) cascades

— continuum (black)

Ref. Ji et al., MNRAS (2006)

Ji et al. (2010) in-preparation

Ref. Ji et al., AIP (2009)

Ji et al. (2010) in-preparation

? AtomDB 2.0.0 will include Innershell ionization database

Ref: Foster et al. 2010 (in-preparation)

? AtomDB 1.3.1 doesn’t have RR/DR cascades

? AtomDB 2.0.0 has such features:

— important in some NEI scenarios

Ref: Foster et al. 2010 (in-preparation)

Reeves et al. (2007)

? Light Curves Comparions for AIA 6 bands

? Ionic Fration Comparison for CIE/NEI with Di�erent Databases

Time (s) Time (s)

Ligh

t Cur

ve (N

orm

aliz

ed U

nits

)

? Solar AIA 6 bands:

— A94: FeXVIII

— A131: FeVIII, FeXX, FeXXII

— A171: FeIX

— A193: FeXII, FeXXIV

— A211: FeXIV

— A335: FeXVI

? Left: Light Curves Comparison

— NEI: solid

— CIE: dash

Ref: Ji et al. (2010) in-preparation

Multi-threaded flare models - NEI effects

Calculate SDO/AIA SDO/EVE spectra/light curve Ji et al. in-preparation


What do we need in the future?


Astro-H Spectroscopic Capabilities Re

solu

tion

x [E

ff. A

rea]

0.5

XMM-Newton RGS

Chandra Gratings

Astro-HSXS

Figure of Merit formeasuring fluxin strong lines

102

103

104

CCDs

1 10Energy (keV)

Astro-H Spectroscopic Capabilities


Line Shapes & Interpretation Fl

ux (c

ount

s/s/

keV)

Energy (keV)

0.5

1.0

6.6 6.65 6.7

Astro-H SXS, 100 ksecPerseus Core (0.5’-1.5’)

No turbulence but with resonant scattering

No turbulence or resonant scattering

Mach 0.5 turb. w/res. scat.

Mach 0.5 turb. w/o res. scat.

Zhuravleva+ 2013

Line Shapes & Interpretation


� �

��

��

��

��

Collaborators: Q. Daniel Wang, J. Kwan, T. Tripp (UMass)

N. Schulz, M. Nowak , H. Marshall, M. Noble ,

T. DeLaney, F.K. Baganoff, Claude R. Canizares (MIT)

T. Kallman (NASA/GSFC)

J. Raymond (CfA)

Ka-Wah Wang, Craig, L. Sarazin, Daniel R. Wik (UGA)

F. Yuan (SAO) � S.N. Zhang, Q.S. Gu (NJU)

Collaborators: A.Foster, R.K. Smith, N.Brickhouse, J. Raymond, K.K. Reeves, P. Testa, A. Engell, E. Deluca (CfA) N. Schulz, M. Nowak, H.Marshall, M.Nobel, J. Houck T. DeLaney, F.K. Baganoff, Claude R. Canizares (MIT) T. Kallman (NASA/GSFC), M. Sun (U. Alabama) Ka-Wah Wang(UAlabam) Craig, L. Sarazin (UGA) Q. Daniel Wang, J. Kwan, T. Tripp, H. Dong (UMass) F. Yuan (SAO), Q.S. Gu, Y. Chen, X. Luo (NJU) J.Chang, S.J. Lei, S.N. Zhang, X. Zhou (PMO)

Thanks


Exploring NEI based on AtomDB 3.0

Documents

Transcript of Exploring NEI based on AtomDB 3.0