Photoionization Modeling: the K Lines and Edges of Iron P. Palmeri (UMH-Belgium) T. Kallman (GSFC/NASA-USA) C. Mendoza & M. Bautista (IVIC-Venezuela) J.

Photoionization Modeling: the K Lines and Edges of Iron

P. Palmeri (UMH-Belgium)

T. Kallman (GSFC/NASA-USA)

C. Mendoza & M. Bautista (IVIC-Venezuela)

J. Krolik (JHU-USA)

Plan

• Introduction

• Atomic Data

• Photoionized Plasma Modeling

• Conclusions

Introduction

• Iron K lines are observed in (almost) all X-ray sources

• First reported in rocket observations of the supernova remnant Cas A

Serlemitsos, et al, 1973

Introduction

• Appear in a relatively unconfused region

• Emitted efficiently over wide range of temperatures and ionization states

• Relativistically broaden and red-shifted lines observed in galactic black hole candidates

Tanaka et al., 1996

Introduction

RXTEEXOSAT

ASCA XMM

Chandra

Astro-E2

Compton

1000 km/s

300 km/s

The world of X-ray observatory is changing:

Atomic Data

• Motivation: they were scarce and not sufficiently accurate especially for the M-shell ions (Fe I-XVII)

• Methods: standard atomic codes, i.e. AUTOSTRUCTURE (Badnell), HFR (Cowan) & BPRM (IP/RmaX Projects)

Atomic Data• L-shell ions (Fe XVIII-XXV) CI: {2s,2p}N+[1s]{2s,2p} N+1+up to double excitations to M-

shellSemi-empirical corrections: compilation of Shirai et al (2000)• M-shell ions (Fe I-XVII)Focus on K-vacancy states produced by removing a 1s

electron from the ground configurationNo experimental energies Ab initio calculationsFew experimental data (wavelengths): Fe X & solid state

Core Relaxation EffectsElectrons in K-vacancy & valence configurations see radically different potentials different orbitals for initial & final states of inner-shell transitions

affects level energies, wavelengths & rates !!!

-increase radiative rates by ~5-10%

-increase KLL rates by ~10%

-no systematic effect on KLM rates

-decrease KMM rates by ~10%

Damping Effect

−+→+ essh NN μμν 11 2

Resonances before K-edge

−− +→ esnps NN 1211 μμ

−− +→ enps N 221 μ

νμ hs N +→ 21

νμ hnps N +→ −121Spectator channels (Damping channels)

Participator channels

Damping Effect: Photoabsorption

Fe XVII Fe XXIII

With damping

Without damping

Damping Effect: Electron Impact

Withoutdamping

Withdamping

Fe XIX

2p4 3P2 [1s]2p5 3Po2 [1s]2p5 3Po

1 [1s]2p5 3Po0

Line Energy vs. Ionization StageBlue=MakishimaBlack=these studies

Line moves to red near Fe IX

Complicated K linestructure

Edge Energy vs. Ionization Stage

In first row ions, ground level is split by various valence configurations

Blue=MakishimaBlack=these studies

In 2nd and 3rd row ions,Splitting is smaller,Results differ significantlyFrom previous

K/K ratio vs. Ionization Stage

MCDF

Auto-S

HFR

Kaastra-Mewe

Jacobs-Rosznyai experiment

K/K ratio is a potential diagnostic of ionization

Fluorescence Yield vs. Ion. Stage

Auto-SHFR

Jacobs-Rosznyai

Kaastra-Mewe

Experiment

Photoionized Plasma Modeling With XSTAR

• Photoionization of a gas by intense external X-ray source (dominant)

• Other processes affecting ionization, excitation & temperature are in equilibrium

Local conditions (ionization fractions, temperature, opacity) parameterized by

=Ionization parameter =4Ionizing flux/gas density

Photoionized Plasma Modeling: Atomic Processes

• Each ion has ~3-30 K-vacancy levels which can be populated by photoionization

• ~4-100 K lines per ion considered in our treatment

Ionization Balance & Temperature

Ionization balance temperature =Ionization parameter=4Ionizing flux/gas density

104<T<108K

Line Emissivity vs.

Emissivity j~ n2

Line Emissivity vs. density

Log =2

Line Emissivity vs. density (continued)

Log(n)=12

Log(n)=16

Line Emissivity vs. Optical Depth Multiplier

Lines can be suppressed by Auger destruction

Line Emissivity vs. Column Density

1/N decrease marks the Breakdown of the optically thinapproximation

Shift of ionization from high to low will be detectable in reprocessed spectrum

Emissivity averaged over constant density slab with log()=2

QuickTime™ and aGIF decompressor

are needed to see this picture.

QuickTime™ and aGIF decompressor

are needed to see this picture.

Simulated Spectra

XMM Epic PN Astro-E XRS

Assuming log()=2, log(N)=23, 100 mcrab source, tobs= 100 ksec

Conclusions

• Structure of Iron K shell is more complicated than has been previously appreciated, & care is needed to accurately compute useful quantities

• There is a shortage of experimental data needed for accurate spectral modeling especially in intermediate & low ionization stages

• Converging series of damped resonances act to smear absorption edges

• Emission lines contain structure which has diagnostic value, even for low ionization gas